{ "format": "renigma.fresh_transfer_summary", "version": 1, "experiment_id": "REP-005", "status": "verified", "attempt": "rep-22de256c2306", "source_commit": "f94db862babf5b122229c929b8c121489fe06e11", "source_sha256": "f80bea899d921731bf1a37b506fec851f8e7245e821522fd33bb56518ec839fc", "receipt_sha256": "dbaaf718f87af0a768af015c4c22ba487d037ccbd23463d5e1b7d855a32eaa0b", "configuration_sha256": "c82ba5de5f4c7d090dcb5af959b6bbd33f0ddcd18a89e41af1be25386cdd312f", "design_sha256": "d4846f73e1789e64744d014f98425046d73b9c5f7cd1ae1384b4130f7122c06b", "started_at": "2026-10-03T22:02:06.268291+00:00", "ended_at": "2026-10-03T22:02:52.476761+00:00", "optimizer_updates": 15000, "worker_seconds": 33.338136700040195, "aggregate_seconds": 46.32799999997951, "training_timing": { "synchronized_wall_seconds": 10.91606060002232, "cuda_event_interval_ms": 10914.9267578125, "scope": "Training interval includes checkpoint and logging gaps; profiler retains actual kernels" }, "evaluation": { "wall_seconds": 15.537679799948819, "families": 41, "trial_state_word_cases": 33558120 }, "gpu": "NVIDIA GeForce RTX 5070", "device_activity": { "samples": 27, "scope": "Device-wide NVIDIA samples, not per-process utilization or speedup evidence", "utilization_percent": { "maximum": 82.0, "sample_mean": 21.51851851851852 }, "power_watts": { "maximum": 145.25, "sample_mean": 61.54666666666667 }, "temperature_celsius": { "maximum": 45.0, "sample_mean": 38.333333333333336 } }, "cuda_memory": { "peak_allocated_bytes": 441035776, "peak_reserved_bytes": 721420288, "device_free_bytes": 10708058112 }, "verification": { "status": "verified", "experiment_id": "REP-005", "mode": "official", "optimizer_updates": 15000, "trials": 20, "source_sha256": "f80bea899d921731bf1a37b506fec851f8e7245e821522fd33bb56518ec839fc", "files_verified": 104, "core": { "status": "verified", "optimizer_updates": 15000, "trials": 20, "world_controls": { "status": "passed", "checks": { "frozen_design": true, "training_rank_10": true, "affine_gauge_rank_10": true, "reference_intertwining": true, "exact_coordinate_roundtrip": true, "all_reference_relations": true, "order_is_observable": true, "erased_context_collides": true, "context_required_for_S": true, "constant_code_fails_observation": true }, "ranks": { "training_features": 10, "training_augmented": 10 }, "reference_tolerance": 1e-12, "reference_max_abs_intertwining_errors": { "X": 1.4561119536544653e-15, "Y": 1.4561119536544653e-15, "Xinv": 1.4432899320127035e-15, "R": 1.4432899320127035e-15, "Rinv": 1.4432899320127035e-15, "S": 1.4561119536544653e-15, "H": 0.0, "P": 0.0 }, "relations": { "X_inverse": true, "inverse_X": true, "R_inverse": true, "inverse_R": true, "X_order_31": true, "Y_order_31": true, "S_order_31": true, "R_order_4": true, "H_order_2": true, "context_inverse": true, "projection_idempotent": true, "rotation_conjugates_translation": true }, "order_sensitive_state_counts": { "translation_rotation": 1922, "context_flip_translation": 1922, "projection_rotation": 1920, "projection_translation": 1922 }, "lost_context_pairs": 961, "constant_code": { "commutator_mse": 0.0, "correct_full_states": 1, "total_states": 1922, "adequate": false }, "scope": "Exact integer truth and binary64 analytic controls; no trained-model result" }, "numpy_replay": { "status": "passed", "families": [ "reconstruction", "individual", "length2", "length4", "length8", "length16", "length32", "length16_no_projection", "length32_no_projection", "relations_X_inverse_left", "relations_X_inverse_right", "relations_inverse_X_left", "relations_inverse_X_right", "relations_R_inverse_left", "relations_R_inverse_right", "relations_inverse_R_left", "relations_inverse_R_right", "relations_X_order_31_left", "relations_X_order_31_right", "relations_Y_order_31_left", "relations_Y_order_31_right", "relations_S_order_31_left", "relations_S_order_31_right", "relations_R_order_4_left", "relations_R_order_4_right", "relations_H_order_2_left", "relations_H_order_2_right", "relations_context_inverse_left", "relations_context_inverse_right", "relations_projection_idempotent_left", "relations_projection_idempotent_right", "relations_rotation_conjugates_translation_left", "relations_rotation_conjugates_translation_right", "noncommuting_translation_rotation_left", "noncommuting_translation_rotation_right", "noncommuting_context_flip_translation_left", "noncommuting_context_flip_translation_right", "noncommuting_projection_rotation_left", "noncommuting_projection_rotation_right", "noncommuting_projection_translation_left", "noncommuting_projection_translation_right" ], "trial_state_word_cases": 279360, "state_ids": [ 0, 127, 254, 381, 508, 635, 762, 889, 1016, 1143, 1270, 1397, 1524, 1651, 1778, 1905 ], "categorical_agreement": "exact for both decoders", "scope": "Independent NumPy inference on every word and every 127th initial state; categorical answers and latent MSE, not full logits; same host, no optimizer replay" }, "profiled_cuda_kernels": 402 }, "scope": "Frozen source/design, complete core records, numerical controls and resources; no retraining" }, "families": { "reconstruction": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "individual": { "trials": [ { "seed": 0, "arm": "native", "cases": 15376, "latent_mse": 0.0024177543818950653, "learned": { "correct": 14947, "accuracy": 0.9720993756503642, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 15376, "latent_mse": 0.000331543677020818, "learned": { "correct": 14955, "accuracy": 0.9726196670135275, "repairs_vs_native": 17, "regressions_vs_native": 9 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 15376, "latent_mse": 0.0003177133039571345, "learned": { "correct": 14957, "accuracy": 0.9727497398543185, "repairs_vs_native": 17, "regressions_vs_native": 7 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 15376, "latent_mse": 0.0003177133039571345, "learned": { "correct": 14957, "accuracy": 0.9727497398543185, "repairs_vs_native": 17, "regressions_vs_native": 7 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 15376, "latent_mse": 0.0021320178639143705, "learned": { "correct": 14846, "accuracy": 0.9655306971904266, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 15376, "latent_mse": 0.0003113021666649729, "learned": { "correct": 14866, "accuracy": 0.966831425598335, "repairs_vs_native": 35, "regressions_vs_native": 15 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 15376, "latent_mse": 0.0003013985697180033, "learned": { "correct": 14864, "accuracy": 0.9667013527575442, "repairs_vs_native": 34, "regressions_vs_native": 16 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 15376, "latent_mse": 0.0003013985697180033, "learned": { "correct": 14864, "accuracy": 0.9667013527575442, "repairs_vs_native": 34, "regressions_vs_native": 16 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 15376, "latent_mse": 0.002170316409319639, "learned": { "correct": 15186, "accuracy": 0.9876430801248699, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 15376, "latent_mse": 0.00024726081755943596, "learned": { "correct": 15183, "accuracy": 0.9874479708636836, "repairs_vs_native": 9, "regressions_vs_native": 12 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 15376, "latent_mse": 0.00023090164177119732, "learned": { "correct": 15182, "accuracy": 0.9873829344432883, "repairs_vs_native": 8, "regressions_vs_native": 12 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 15376, "latent_mse": 0.00023090164177119732, "learned": { "correct": 15182, "accuracy": 0.9873829344432883, "repairs_vs_native": 8, "regressions_vs_native": 12 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 15376, "latent_mse": 0.0023816442117094994, "learned": { "correct": 15201, "accuracy": 0.9886186264308012, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 15376, "latent_mse": 0.0003855224931612611, "learned": { "correct": 15255, "accuracy": 0.992130593132154, "repairs_vs_native": 60, "regressions_vs_native": 6 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 15376, "latent_mse": 0.0003702655085362494, "learned": { "correct": 15257, "accuracy": 0.9922606659729448, "repairs_vs_native": 61, "regressions_vs_native": 5 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 15376, "latent_mse": 0.0003702655085362494, "learned": { "correct": 15257, "accuracy": 0.9922606659729448, "repairs_vs_native": 61, "regressions_vs_native": 5 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 15376, "latent_mse": 0.002433223184198141, "learned": { "correct": 14987, "accuracy": 0.9747008324661811, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 15376, "latent_mse": 0.00044029319542460144, "learned": { "correct": 15033, "accuracy": 0.9776925078043704, "repairs_vs_native": 57, "regressions_vs_native": 11 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 15376, "latent_mse": 0.00042737717740237713, "learned": { "correct": 15031, "accuracy": 0.9775624349635796, "repairs_vs_native": 56, "regressions_vs_native": 12 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 15376, "latent_mse": 0.00042737717740237713, "learned": { "correct": 15031, "accuracy": 0.9775624349635796, "repairs_vs_native": 56, "regressions_vs_native": 12 }, "reference": { "correct": 15376, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9747008324661811, "minimum_accuracy": 0.9655306971904266, "maximum_accuracy": 0.9886186264308012, "cases": 76880, "correct": 75167, "errors": 1713, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9776925078043704, "minimum_accuracy": 0.966831425598335, "maximum_accuracy": 0.992130593132154, "cases": 76880, "correct": 75292, "errors": 1588, "repairs_vs_native": 178, "regressions_vs_native": 53, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9775624349635796, "minimum_accuracy": 0.9667013527575442, "maximum_accuracy": 0.9922606659729448, "cases": 76880, "correct": 75291, "errors": 1589, "repairs_vs_native": 176, "regressions_vs_native": 52, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9775624349635796, "minimum_accuracy": 0.9667013527575442, "maximum_accuracy": 0.9922606659729448, "cases": 76880, "correct": 75291, "errors": 1589, "repairs_vs_native": 176, "regressions_vs_native": 52, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 76880, "correct": 76880, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 76880, "correct": 76880, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 76880, "correct": 76880, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 76880, "correct": 76880, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "length2": { "trials": [ { "seed": 0, "arm": "native", "cases": 123008, "latent_mse": 0.005189226474612951, "learned": { "correct": 116872, "accuracy": 0.9501170655567117, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 123008, "latent_mse": 0.0008288945537060499, "learned": { "correct": 117209, "accuracy": 0.9528567247658689, "repairs_vs_native": 543, "regressions_vs_native": 206 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 123008, "latent_mse": 0.0007553204777650535, "learned": { "correct": 117269, "accuracy": 0.9533444979188346, "repairs_vs_native": 557, "regressions_vs_native": 160 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 123008, "latent_mse": 0.0007551456801593304, "learned": { "correct": 117270, "accuracy": 0.9533526274713839, "repairs_vs_native": 557, "regressions_vs_native": 159 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 123008, "latent_mse": 0.004528135526925325, "learned": { "correct": 114801, "accuracy": 0.9332807622268471, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 123008, "latent_mse": 0.0007454268052242696, "learned": { "correct": 115937, "accuracy": 0.9425159339229969, "repairs_vs_native": 1313, "regressions_vs_native": 177 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 123008, "latent_mse": 0.0006932102842256427, "learned": { "correct": 116056, "accuracy": 0.9434833506763788, "repairs_vs_native": 1480, "regressions_vs_native": 225 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 123008, "latent_mse": 0.0006928169750608504, "learned": { "correct": 116085, "accuracy": 0.9437191077003122, "repairs_vs_native": 1509, "regressions_vs_native": 225 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 123008, "latent_mse": 0.004674298223108053, "learned": { "correct": 119015, "accuracy": 0.9675386966701353, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 123008, "latent_mse": 0.000685665407218039, "learned": { "correct": 119163, "accuracy": 0.9687418704474505, "repairs_vs_native": 317, "regressions_vs_native": 169 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 123008, "latent_mse": 0.0006005785544402897, "learned": { "correct": 119155, "accuracy": 0.9686768340270552, "repairs_vs_native": 321, "regressions_vs_native": 181 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 123008, "latent_mse": 0.0006004657479934394, "learned": { "correct": 119161, "accuracy": 0.9687256113423517, "repairs_vs_native": 321, "regressions_vs_native": 175 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 123008, "latent_mse": 0.005122740752995014, "learned": { "correct": 117968, "accuracy": 0.9590270551508845, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 123008, "latent_mse": 0.0009902110323309898, "learned": { "correct": 118408, "accuracy": 0.9626040582726326, "repairs_vs_native": 990, "regressions_vs_native": 550 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 123008, "latent_mse": 0.0009092063992284238, "learned": { "correct": 118454, "accuracy": 0.9629780176899063, "repairs_vs_native": 1054, "regressions_vs_native": 568 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 123008, "latent_mse": 0.000909080496057868, "learned": { "correct": 118487, "accuracy": 0.9632462929240374, "repairs_vs_native": 1055, "regressions_vs_native": 536 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 123008, "latent_mse": 0.00513590918853879, "learned": { "correct": 116326, "accuracy": 0.9456783298647242, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 123008, "latent_mse": 0.0010730347130447626, "learned": { "correct": 117314, "accuracy": 0.9537103277835588, "repairs_vs_native": 1175, "regressions_vs_native": 187 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 123008, "latent_mse": 0.0009983180789276958, "learned": { "correct": 117292, "accuracy": 0.9535314776274714, "repairs_vs_native": 1182, "regressions_vs_native": 216 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 123008, "latent_mse": 0.0009959900053218007, "learned": { "correct": 117318, "accuracy": 0.9537428459937565, "repairs_vs_native": 1181, "regressions_vs_native": 189 }, "reference": { "correct": 123008, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9501170655567117, "minimum_accuracy": 0.9332807622268471, "maximum_accuracy": 0.9675386966701353, "cases": 615040, "correct": 584982, "errors": 30058, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9537103277835588, "minimum_accuracy": 0.9425159339229969, "maximum_accuracy": 0.9687418704474505, "cases": 615040, "correct": 588031, "errors": 27009, "repairs_vs_native": 4338, "regressions_vs_native": 1289, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9535314776274714, "minimum_accuracy": 0.9434833506763788, "maximum_accuracy": 0.9686768340270552, "cases": 615040, "correct": 588226, "errors": 26814, "repairs_vs_native": 4594, "regressions_vs_native": 1350, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9537428459937565, "minimum_accuracy": 0.9437191077003122, "maximum_accuracy": 0.9687256113423517, "cases": 615040, "correct": 588321, "errors": 26719, "repairs_vs_native": 4623, "regressions_vs_native": 1284, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 615040, "correct": 615040, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 615040, "correct": 615040, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 615040, "correct": 615040, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 615040, "correct": 615040, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "length4": { "trials": [ { "seed": 0, "arm": "native", "cases": 246016, "latent_mse": 0.009888578206300735, "learned": { "correct": 230033, "accuracy": 0.9350326808012487, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 245956, "accuracy": 0.9997561134235172, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.002251700032502413, "learned": { "correct": 233499, "accuracy": 0.9491211953694069, "repairs_vs_native": 4311, "regressions_vs_native": 845 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 60, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 246016, "latent_mse": 0.0019547343254089355, "learned": { "correct": 233586, "accuracy": 0.9494748309053069, "repairs_vs_native": 4310, "regressions_vs_native": 757 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 60, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 246016, "latent_mse": 0.001953711034730077, "learned": { "correct": 233586, "accuracy": 0.9494748309053069, "repairs_vs_native": 4312, "regressions_vs_native": 759 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 60, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 246016, "latent_mse": 0.008413389325141907, "learned": { "correct": 221811, "accuracy": 0.9016120902705516, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.0018387897871434689, "learned": { "correct": 226873, "accuracy": 0.9221879877731529, "repairs_vs_native": 6220, "regressions_vs_native": 1158 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 246016, "latent_mse": 0.0016317380359396338, "learned": { "correct": 226579, "accuracy": 0.9209929435483871, "repairs_vs_native": 6119, "regressions_vs_native": 1351 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 246016, "latent_mse": 0.0016292689833790064, "learned": { "correct": 226486, "accuracy": 0.9206149193548387, "repairs_vs_native": 6100, "regressions_vs_native": 1425 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 246016, "latent_mse": 0.008990274742245674, "learned": { "correct": 231230, "accuracy": 0.9398982180020812, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.002089225221425295, "learned": { "correct": 230851, "accuracy": 0.9383576677939646, "repairs_vs_native": 1728, "regressions_vs_native": 2107 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 246016, "latent_mse": 0.001741556217893958, "learned": { "correct": 230882, "accuracy": 0.9384836758584808, "repairs_vs_native": 1759, "regressions_vs_native": 2107 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 246016, "latent_mse": 0.001741071348078549, "learned": { "correct": 231012, "accuracy": 0.9390120967741935, "repairs_vs_native": 1759, "regressions_vs_native": 1977 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 246016, "latent_mse": 0.009816489182412624, "learned": { "correct": 229902, "accuracy": 0.9345001951092612, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.0026377560570836067, "learned": { "correct": 232226, "accuracy": 0.9439467351716961, "repairs_vs_native": 3563, "regressions_vs_native": 1239 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 246016, "latent_mse": 0.0023065877612680197, "learned": { "correct": 232417, "accuracy": 0.9447231074401665, "repairs_vs_native": 3668, "regressions_vs_native": 1153 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 246016, "latent_mse": 0.0023058964870870113, "learned": { "correct": 232438, "accuracy": 0.9448084677419355, "repairs_vs_native": 3674, "regressions_vs_native": 1138 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 246016, "latent_mse": 0.009486116468906403, "learned": { "correct": 226580, "accuracy": 0.9209970083246618, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.002696767682209611, "learned": { "correct": 230955, "accuracy": 0.9387804045265349, "repairs_vs_native": 5165, "regressions_vs_native": 790 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 246016, "latent_mse": 0.0023829417768865824, "learned": { "correct": 230900, "accuracy": 0.9385568418314256, "repairs_vs_native": 5173, "regressions_vs_native": 853 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 246016, "latent_mse": 0.002368774265050888, "learned": { "correct": 230922, "accuracy": 0.9386462669094693, "repairs_vs_native": 5196, "regressions_vs_native": 854 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9345001951092612, "minimum_accuracy": 0.9016120902705516, "maximum_accuracy": 0.9398982180020812, "cases": 1230080, "correct": 1139556, "errors": 90524, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9387804045265349, "minimum_accuracy": 0.9221879877731529, "maximum_accuracy": 0.9491211953694069, "cases": 1230080, "correct": 1154404, "errors": 75676, "repairs_vs_native": 20987, "regressions_vs_native": 6139, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9385568418314256, "minimum_accuracy": 0.9209929435483871, "maximum_accuracy": 0.9494748309053069, "cases": 1230080, "correct": 1154364, "errors": 75716, "repairs_vs_native": 21029, "regressions_vs_native": 6221, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9390120967741935, "minimum_accuracy": 0.9206149193548387, "maximum_accuracy": 0.9494748309053069, "cases": 1230080, "correct": 1154444, "errors": 75636, "repairs_vs_native": 21041, "regressions_vs_native": 6153, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 0.9997561134235172, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1230020, "errors": 60, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1230080, "errors": 0, "repairs_vs_native": 60, "regressions_vs_native": 0, "regression_free_recovery": true, "perfect_recovery": true }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1230080, "errors": 0, "repairs_vs_native": 60, "regressions_vs_native": 0, "regression_free_recovery": true, "perfect_recovery": true }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1230080, "errors": 0, "repairs_vs_native": 60, "regressions_vs_native": 0, "regression_free_recovery": true, "perfect_recovery": true } } } }, "length8": { "trials": [ { "seed": 0, "arm": "native", "cases": 246016, "latent_mse": 0.031227294355630875, "learned": { "correct": 205491, "accuracy": 0.8352749414672217, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 245959, "accuracy": 0.9997683077523413, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.007110559847205877, "learned": { "correct": 221229, "accuracy": 0.8992463904786681, "repairs_vs_native": 18251, "regressions_vs_native": 2513 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 57, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 246016, "latent_mse": 0.005590937100350857, "learned": { "correct": 221427, "accuracy": 0.9000512161810614, "repairs_vs_native": 17966, "regressions_vs_native": 2030 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 57, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 246016, "latent_mse": 0.005582100711762905, "learned": { "correct": 221351, "accuracy": 0.8997422931841832, "repairs_vs_native": 17888, "regressions_vs_native": 2028 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 57, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 246016, "latent_mse": 0.026762278750538826, "learned": { "correct": 192267, "accuracy": 0.7815223400104059, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 245957, "accuracy": 0.9997601781997919, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.005846222396939993, "learned": { "correct": 206955, "accuracy": 0.8412257739334027, "repairs_vs_native": 21282, "regressions_vs_native": 6594 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 59, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 246016, "latent_mse": 0.004792678635567427, "learned": { "correct": 208313, "accuracy": 0.8467457401144641, "repairs_vs_native": 22156, "regressions_vs_native": 6110 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 59, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 246016, "latent_mse": 0.004768306855112314, "learned": { "correct": 208861, "accuracy": 0.8489732375130072, "repairs_vs_native": 22406, "regressions_vs_native": 5812 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 59, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 246016, "latent_mse": 0.02896590158343315, "learned": { "correct": 213436, "accuracy": 0.8675695889698231, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 245814, "accuracy": 0.9991789151925078, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.0072002955712378025, "learned": { "correct": 219066, "accuracy": 0.890454279396462, "repairs_vs_native": 10580, "regressions_vs_native": 4950 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 202, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 246016, "latent_mse": 0.005439284723252058, "learned": { "correct": 219572, "accuracy": 0.8925110561914672, "repairs_vs_native": 10732, "regressions_vs_native": 4596 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 202, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 246016, "latent_mse": 0.005434753838926554, "learned": { "correct": 219678, "accuracy": 0.8929419224765869, "repairs_vs_native": 10738, "regressions_vs_native": 4496 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 202, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 246016, "latent_mse": 0.031029842793941498, "learned": { "correct": 206635, "accuracy": 0.8399250455254943, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 242918, "accuracy": 0.9874073231009365, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.008530835621058941, "learned": { "correct": 211414, "accuracy": 0.8593506113423517, "repairs_vs_native": 11334, "regressions_vs_native": 6555 }, "reference": { "correct": 243482, "accuracy": 0.9896998569198752, "repairs_vs_native": 1289, "regressions_vs_native": 725 } }, { "seed": 3, "arm": "latent_only", "cases": 246016, "latent_mse": 0.006872852798551321, "learned": { "correct": 211753, "accuracy": 0.8607285704994797, "repairs_vs_native": 11888, "regressions_vs_native": 6770 }, "reference": { "correct": 243443, "accuracy": 0.9895413306451613, "repairs_vs_native": 1243, "regressions_vs_native": 718 } }, { "seed": 3, "arm": "refit", "cases": 246016, "latent_mse": 0.00686693424358964, "learned": { "correct": 211973, "accuracy": 0.8616228212799167, "repairs_vs_native": 11962, "regressions_vs_native": 6624 }, "reference": { "correct": 243451, "accuracy": 0.989573848855359, "repairs_vs_native": 1253, "regressions_vs_native": 720 } }, { "seed": 4, "arm": "native", "cases": 246016, "latent_mse": 0.030039425939321518, "learned": { "correct": 196304, "accuracy": 0.7979318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 244874, "accuracy": 0.9953580254942768, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.008374999277293682, "learned": { "correct": 212426, "accuracy": 0.8634641649323621, "repairs_vs_native": 17915, "regressions_vs_native": 1793 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 1142, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 246016, "latent_mse": 0.006893188692629337, "learned": { "correct": 213124, "accuracy": 0.8663013787721123, "repairs_vs_native": 18587, "regressions_vs_native": 1767 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 1142, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 246016, "latent_mse": 0.006789231672883034, "learned": { "correct": 214566, "accuracy": 0.8721627861602498, "repairs_vs_native": 19719, "regressions_vs_native": 1457 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 1142, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.8352749414672217, "minimum_accuracy": 0.7815223400104059, "maximum_accuracy": 0.8675695889698231, "cases": 1230080, "correct": 1014133, "errors": 215947, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.8634641649323621, "minimum_accuracy": 0.8412257739334027, "maximum_accuracy": 0.8992463904786681, "cases": 1230080, "correct": 1071090, "errors": 158990, "repairs_vs_native": 79362, "regressions_vs_native": 22405, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.8663013787721123, "minimum_accuracy": 0.8467457401144641, "maximum_accuracy": 0.9000512161810614, "cases": 1230080, "correct": 1074189, "errors": 155891, "repairs_vs_native": 81329, "regressions_vs_native": 21273, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.8721627861602498, "minimum_accuracy": 0.8489732375130072, "maximum_accuracy": 0.8997422931841832, "cases": 1230080, "correct": 1076429, "errors": 153651, "repairs_vs_native": 82713, "regressions_vs_native": 20417, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 0.9991789151925078, "minimum_accuracy": 0.9874073231009365, "maximum_accuracy": 0.9997683077523413, "cases": 1230080, "correct": 1225522, "errors": 4558, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 0.9896998569198752, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1227546, "errors": 2534, "repairs_vs_native": 2749, "regressions_vs_native": 725, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 0.9895413306451613, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1227507, "errors": 2573, "repairs_vs_native": 2703, "regressions_vs_native": 718, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 0.989573848855359, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1227515, "errors": 2565, "repairs_vs_native": 2713, "regressions_vs_native": 720, "regression_free_recovery": false, "perfect_recovery": false } } } }, "length16": { "trials": [ { "seed": 0, "arm": "native", "cases": 246016, "latent_mse": 0.07042398303747177, "learned": { "correct": 160905, "accuracy": 0.6540428264828304, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 224850, "accuracy": 0.9139649453694069, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.02046845853328705, "learned": { "correct": 204669, "accuracy": 0.8319336953694069, "repairs_vs_native": 51058, "regressions_vs_native": 7294 }, "reference": { "correct": 245835, "accuracy": 0.9992642754942768, "repairs_vs_native": 20985, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 246016, "latent_mse": 0.016099393367767334, "learned": { "correct": 206806, "accuracy": 0.8406201222684704, "repairs_vs_native": 52546, "regressions_vs_native": 6645 }, "reference": { "correct": 245844, "accuracy": 0.9993008584807492, "repairs_vs_native": 20994, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 246016, "latent_mse": 0.016059987246990204, "learned": { "correct": 206953, "accuracy": 0.8412176443808532, "repairs_vs_native": 52623, "regressions_vs_native": 6575 }, "reference": { "correct": 245843, "accuracy": 0.9992967937044746, "repairs_vs_native": 20993, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 246016, "latent_mse": 0.05873805657029152, "learned": { "correct": 156127, "accuracy": 0.6346213254422477, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 232909, "accuracy": 0.9467229773673257, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.01706366240978241, "learned": { "correct": 199414, "accuracy": 0.8105732960457857, "repairs_vs_native": 52515, "regressions_vs_native": 9228 }, "reference": { "correct": 245889, "accuracy": 0.9994837734131113, "repairs_vs_native": 12980, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 246016, "latent_mse": 0.014042847789824009, "learned": { "correct": 199084, "accuracy": 0.8092319198751301, "repairs_vs_native": 52741, "regressions_vs_native": 9784 }, "reference": { "correct": 245889, "accuracy": 0.9994837734131113, "repairs_vs_native": 12980, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 246016, "latent_mse": 0.013939269818365574, "learned": { "correct": 198572, "accuracy": 0.8071507544224766, "repairs_vs_native": 52716, "regressions_vs_native": 10271 }, "reference": { "correct": 245889, "accuracy": 0.9994837734131113, "repairs_vs_native": 12980, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 246016, "latent_mse": 0.06610509008169174, "learned": { "correct": 183025, "accuracy": 0.7439556776795005, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 232593, "accuracy": 0.9454385080645161, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.021809278056025505, "learned": { "correct": 204909, "accuracy": 0.8329092416753382, "repairs_vs_native": 34007, "regressions_vs_native": 12123 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 13423, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 246016, "latent_mse": 0.01660114713013172, "learned": { "correct": 207301, "accuracy": 0.8426321865244537, "repairs_vs_native": 35311, "regressions_vs_native": 11035 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 13423, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 246016, "latent_mse": 0.016589175909757614, "learned": { "correct": 207383, "accuracy": 0.8429654981789803, "repairs_vs_native": 35352, "regressions_vs_native": 10994 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 13423, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 246016, "latent_mse": 0.0703839361667633, "learned": { "correct": 172081, "accuracy": 0.6994707661290323, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 219942, "accuracy": 0.8940150234131113, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.024510890245437622, "learned": { "correct": 193228, "accuracy": 0.7854285900104059, "repairs_vs_native": 31627, "regressions_vs_native": 10480 }, "reference": { "correct": 236780, "accuracy": 0.962457726326743, "repairs_vs_native": 20152, "regressions_vs_native": 3314 } }, { "seed": 3, "arm": "latent_only", "cases": 246016, "latent_mse": 0.01982893981039524, "learned": { "correct": 195536, "accuracy": 0.7948100936524454, "repairs_vs_native": 32580, "regressions_vs_native": 9125 }, "reference": { "correct": 236608, "accuracy": 0.9617585848074922, "repairs_vs_native": 20102, "regressions_vs_native": 3436 } }, { "seed": 3, "arm": "refit", "cases": 246016, "latent_mse": 0.01980399712920189, "learned": { "correct": 195577, "accuracy": 0.7949767494797086, "repairs_vs_native": 32610, "regressions_vs_native": 9114 }, "reference": { "correct": 236592, "accuracy": 0.9616935483870968, "repairs_vs_native": 20120, "regressions_vs_native": 3470 } }, { "seed": 4, "arm": "native", "cases": 246016, "latent_mse": 0.06466652452945709, "learned": { "correct": 152327, "accuracy": 0.619175175598335, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 229896, "accuracy": 0.9344758064516129, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.02401820756494999, "learned": { "correct": 187480, "accuracy": 0.7620642559833507, "repairs_vs_native": 38617, "regressions_vs_native": 3464 }, "reference": { "correct": 245043, "accuracy": 0.9960449726847035, "repairs_vs_native": 15147, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 246016, "latent_mse": 0.019696854054927826, "learned": { "correct": 189140, "accuracy": 0.7688117845993756, "repairs_vs_native": 39708, "regressions_vs_native": 2895 }, "reference": { "correct": 245043, "accuracy": 0.9960449726847035, "repairs_vs_native": 15147, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 246016, "latent_mse": 0.019413534551858902, "learned": { "correct": 191604, "accuracy": 0.7788273933402705, "repairs_vs_native": 41984, "regressions_vs_native": 2707 }, "reference": { "correct": 245043, "accuracy": 0.9960449726847035, "repairs_vs_native": 15147, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.6540428264828304, "minimum_accuracy": 0.619175175598335, "maximum_accuracy": 0.7439556776795005, "cases": 1230080, "correct": 824465, "errors": 405615, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.8105732960457857, "minimum_accuracy": 0.7620642559833507, "maximum_accuracy": 0.8329092416753382, "cases": 1230080, "correct": 989700, "errors": 240380, "repairs_vs_native": 207824, "regressions_vs_native": 42589, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.8092319198751301, "minimum_accuracy": 0.7688117845993756, "maximum_accuracy": 0.8426321865244537, "cases": 1230080, "correct": 997867, "errors": 232213, "repairs_vs_native": 212886, "regressions_vs_native": 39484, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.8071507544224766, "minimum_accuracy": 0.7788273933402705, "maximum_accuracy": 0.8429654981789803, "cases": 1230080, "correct": 1000089, "errors": 229991, "repairs_vs_native": 215285, "regressions_vs_native": 39661, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 0.9344758064516129, "minimum_accuracy": 0.8940150234131113, "maximum_accuracy": 0.9467229773673257, "cases": 1230080, "correct": 1140190, "errors": 89890, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9992642754942768, "minimum_accuracy": 0.962457726326743, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1219563, "errors": 10517, "repairs_vs_native": 82687, "regressions_vs_native": 3314, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9993008584807492, "minimum_accuracy": 0.9617585848074922, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1219400, "errors": 10680, "repairs_vs_native": 82646, "regressions_vs_native": 3436, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9992967937044746, "minimum_accuracy": 0.9616935483870968, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1219383, "errors": 10697, "repairs_vs_native": 82663, "regressions_vs_native": 3470, "regression_free_recovery": false, "perfect_recovery": false } } } }, "length32": { "trials": [ { "seed": 0, "arm": "native", "cases": 246016, "latent_mse": 0.17764726281166077, "learned": { "correct": 78791, "accuracy": 0.32026778746097817, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 148066, "accuracy": 0.6018551638917794, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.055189099162817, "learned": { "correct": 189381, "accuracy": 0.7697913956815817, "repairs_vs_native": 117138, "regressions_vs_native": 6548 }, "reference": { "correct": 245714, "accuracy": 0.9987724375650364, "repairs_vs_native": 97919, "regressions_vs_native": 271 } }, { "seed": 0, "arm": "latent_only", "cases": 246016, "latent_mse": 0.041116863489151, "learned": { "correct": 194615, "accuracy": 0.791066434703434, "repairs_vs_native": 121355, "regressions_vs_native": 5531 }, "reference": { "correct": 245615, "accuracy": 0.9983700247138397, "repairs_vs_native": 97919, "regressions_vs_native": 370 } }, { "seed": 0, "arm": "refit", "cases": 246016, "latent_mse": 0.040849924087524414, "learned": { "correct": 194836, "accuracy": 0.7919647502601457, "repairs_vs_native": 121743, "regressions_vs_native": 5698 }, "reference": { "correct": 245650, "accuracy": 0.9985122918834547, "repairs_vs_native": 97919, "regressions_vs_native": 335 } }, { "seed": 1, "arm": "native", "cases": 246016, "latent_mse": 0.1490212082862854, "learned": { "correct": 79448, "accuracy": 0.3229383454734651, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 163675, "accuracy": 0.6653022567637877, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.048223379999399185, "learned": { "correct": 180616, "accuracy": 0.7341636316337149, "repairs_vs_native": 109083, "regressions_vs_native": 7915 }, "reference": { "correct": 244817, "accuracy": 0.9951263332466181, "repairs_vs_native": 81142, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 246016, "latent_mse": 0.03860945999622345, "learned": { "correct": 181991, "accuracy": 0.7397526990114464, "repairs_vs_native": 109951, "regressions_vs_native": 7408 }, "reference": { "correct": 245199, "accuracy": 0.9966790777835588, "repairs_vs_native": 81524, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 246016, "latent_mse": 0.037900034338235855, "learned": { "correct": 179319, "accuracy": 0.728891616805411, "repairs_vs_native": 109019, "regressions_vs_native": 9148 }, "reference": { "correct": 245353, "accuracy": 0.9973050533298647, "repairs_vs_native": 81678, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 246016, "latent_mse": 0.17078402638435364, "learned": { "correct": 106172, "accuracy": 0.4315654266389178, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 153482, "accuracy": 0.6238699921956296, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.06092651188373566, "learned": { "correct": 168886, "accuracy": 0.6864838059313215, "repairs_vs_native": 85613, "regressions_vs_native": 22899 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 92534, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 246016, "latent_mse": 0.04427139088511467, "learned": { "correct": 172805, "accuracy": 0.7024136641519251, "repairs_vs_native": 87025, "regressions_vs_native": 20392 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 92534, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 246016, "latent_mse": 0.04421594738960266, "learned": { "correct": 172811, "accuracy": 0.7024380528095734, "repairs_vs_native": 87128, "regressions_vs_native": 20489 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 92534, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 246016, "latent_mse": 0.18005837500095367, "learned": { "correct": 111385, "accuracy": 0.45275510535900104, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 136521, "accuracy": 0.5549273218002081, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.06501193344593048, "learned": { "correct": 168904, "accuracy": 0.6865569719042663, "repairs_vs_native": 74780, "regressions_vs_native": 17261 }, "reference": { "correct": 230129, "accuracy": 0.9354228993236212, "repairs_vs_native": 96914, "regressions_vs_native": 3306 } }, { "seed": 3, "arm": "latent_only", "cases": 246016, "latent_mse": 0.04982250928878784, "learned": { "correct": 175437, "accuracy": 0.7131121553069719, "repairs_vs_native": 78632, "regressions_vs_native": 14580 }, "reference": { "correct": 229671, "accuracy": 0.9335612317898023, "repairs_vs_native": 96295, "regressions_vs_native": 3145 } }, { "seed": 3, "arm": "refit", "cases": 246016, "latent_mse": 0.04967866465449333, "learned": { "correct": 175465, "accuracy": 0.7132259690426639, "repairs_vs_native": 78789, "regressions_vs_native": 14709 }, "reference": { "correct": 229770, "accuracy": 0.933963644640999, "repairs_vs_native": 96321, "regressions_vs_native": 3072 } }, { "seed": 4, "arm": "native", "cases": 246016, "latent_mse": 0.16177576780319214, "learned": { "correct": 81956, "accuracy": 0.33313280437044746, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 149719, "accuracy": 0.6085742390738814, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.06034791097044945, "learned": { "correct": 153639, "accuracy": 0.6245081620707597, "repairs_vs_native": 76772, "regressions_vs_native": 5089 }, "reference": { "correct": 243131, "accuracy": 0.9882731204474505, "repairs_vs_native": 93412, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 246016, "latent_mse": 0.04758007079362869, "learned": { "correct": 156312, "accuracy": 0.6353733090530698, "repairs_vs_native": 78904, "regressions_vs_native": 4548 }, "reference": { "correct": 243314, "accuracy": 0.9890169745057232, "repairs_vs_native": 93595, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 246016, "latent_mse": 0.04696023836731911, "learned": { "correct": 166057, "accuracy": 0.6749845538501561, "repairs_vs_native": 87412, "regressions_vs_native": 3311 }, "reference": { "correct": 242777, "accuracy": 0.9868341896462018, "repairs_vs_native": 93058, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.33313280437044746, "minimum_accuracy": 0.32026778746097817, "maximum_accuracy": 0.45275510535900104, "cases": 1230080, "correct": 457752, "errors": 772328, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.6865569719042663, "minimum_accuracy": 0.6245081620707597, "maximum_accuracy": 0.7697913956815817, "cases": 1230080, "correct": 861426, "errors": 368654, "repairs_vs_native": 463386, "regressions_vs_native": 59712, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.7131121553069719, "minimum_accuracy": 0.6353733090530698, "maximum_accuracy": 0.791066434703434, "cases": 1230080, "correct": 881160, "errors": 348920, "repairs_vs_native": 475867, "regressions_vs_native": 52459, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.7132259690426639, "minimum_accuracy": 0.6749845538501561, "maximum_accuracy": 0.7919647502601457, "cases": 1230080, "correct": 888488, "errors": 341592, "repairs_vs_native": 484091, "regressions_vs_native": 53355, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 0.6085742390738814, "minimum_accuracy": 0.5549273218002081, "maximum_accuracy": 0.6653022567637877, "cases": 1230080, "correct": 751463, "errors": 478617, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9951263332466181, "minimum_accuracy": 0.9354228993236212, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1209807, "errors": 20273, "repairs_vs_native": 461921, "regressions_vs_native": 3577, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9966790777835588, "minimum_accuracy": 0.9335612317898023, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1209815, "errors": 20265, "repairs_vs_native": 461867, "regressions_vs_native": 3515, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9973050533298647, "minimum_accuracy": 0.933963644640999, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1209566, "errors": 20514, "repairs_vs_native": 461510, "regressions_vs_native": 3407, "regression_free_recovery": false, "perfect_recovery": false } } } }, "length16_no_projection": { "trials": [ { "seed": 0, "arm": "native", "cases": 246016, "latent_mse": 0.055974479764699936, "learned": { "correct": 195153, "accuracy": 0.7932532843392299, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 235655, "accuracy": 0.9578848530176899, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.03521334007382393, "learned": { "correct": 230036, "accuracy": 0.9350448751300728, "repairs_vs_native": 39028, "regressions_vs_native": 4145 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 10361, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 246016, "latent_mse": 0.026514606550335884, "learned": { "correct": 232967, "accuracy": 0.9469587343912591, "repairs_vs_native": 40608, "regressions_vs_native": 2794 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 10361, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 246016, "latent_mse": 0.026501266285777092, "learned": { "correct": 232941, "accuracy": 0.9468530502081165, "repairs_vs_native": 40613, "regressions_vs_native": 2825 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 10361, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 246016, "latent_mse": 0.04829947277903557, "learned": { "correct": 184620, "accuracy": 0.750438995837669, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 240167, "accuracy": 0.9762251235691988, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.021879032254219055, "learned": { "correct": 226109, "accuracy": 0.9190824986992716, "repairs_vs_native": 46989, "regressions_vs_native": 5500 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 5849, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 246016, "latent_mse": 0.016539353877305984, "learned": { "correct": 229203, "accuracy": 0.9316589164932362, "repairs_vs_native": 48767, "regressions_vs_native": 4184 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 5849, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 246016, "latent_mse": 0.016529595479369164, "learned": { "correct": 229200, "accuracy": 0.931646722164412, "repairs_vs_native": 48770, "regressions_vs_native": 4190 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 5849, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 246016, "latent_mse": 0.05365605652332306, "learned": { "correct": 209496, "accuracy": 0.8515543704474505, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 238524, "accuracy": 0.9695466961498439, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.036296699196100235, "learned": { "correct": 227676, "accuracy": 0.9254520031217481, "repairs_vs_native": 24760, "regressions_vs_native": 6580 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 7492, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 246016, "latent_mse": 0.026696331799030304, "learned": { "correct": 231445, "accuracy": 0.9407721449011447, "repairs_vs_native": 26335, "regressions_vs_native": 4386 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 7492, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 246016, "latent_mse": 0.026676150038838387, "learned": { "correct": 231349, "accuracy": 0.9403819263787722, "repairs_vs_native": 26321, "regressions_vs_native": 4468 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 7492, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 246016, "latent_mse": 0.05322757735848427, "learned": { "correct": 198024, "accuracy": 0.8049232570239334, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 236034, "accuracy": 0.9594254032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.03670807182788849, "learned": { "correct": 219517, "accuracy": 0.8922874934963579, "repairs_vs_native": 29201, "regressions_vs_native": 7708 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 9982, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 246016, "latent_mse": 0.027504904195666313, "learned": { "correct": 224128, "accuracy": 0.9110301768990635, "repairs_vs_native": 31349, "regressions_vs_native": 5245 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 9982, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 246016, "latent_mse": 0.027481039986014366, "learned": { "correct": 224161, "accuracy": 0.911164314516129, "repairs_vs_native": 31442, "regressions_vs_native": 5305 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 9982, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 246016, "latent_mse": 0.05319885164499283, "learned": { "correct": 185379, "accuracy": 0.7535241610301769, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 231660, "accuracy": 0.9416460718002081, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.02798796258866787, "learned": { "correct": 222423, "accuracy": 0.9040997333506764, "repairs_vs_native": 40919, "regressions_vs_native": 3875 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 14356, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 246016, "latent_mse": 0.02110086940228939, "learned": { "correct": 225544, "accuracy": 0.9167859001040582, "repairs_vs_native": 42929, "regressions_vs_native": 2764 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 14356, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 246016, "latent_mse": 0.021061668172478676, "learned": { "correct": 225083, "accuracy": 0.9149120382414152, "repairs_vs_native": 42675, "regressions_vs_native": 2971 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 14356, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.7932532843392299, "minimum_accuracy": 0.750438995837669, "maximum_accuracy": 0.8515543704474505, "cases": 1230080, "correct": 972672, "errors": 257408, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9190824986992716, "minimum_accuracy": 0.8922874934963579, "maximum_accuracy": 0.9350448751300728, "cases": 1230080, "correct": 1125761, "errors": 104319, "repairs_vs_native": 180897, "regressions_vs_native": 27808, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9316589164932362, "minimum_accuracy": 0.9110301768990635, "maximum_accuracy": 0.9469587343912591, "cases": 1230080, "correct": 1143287, "errors": 86793, "repairs_vs_native": 189988, "regressions_vs_native": 19373, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.931646722164412, "minimum_accuracy": 0.911164314516129, "maximum_accuracy": 0.9468530502081165, "cases": 1230080, "correct": 1142734, "errors": 87346, "repairs_vs_native": 189821, "regressions_vs_native": 19759, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 0.9594254032258065, "minimum_accuracy": 0.9416460718002081, "maximum_accuracy": 0.9762251235691988, "cases": 1230080, "correct": 1182040, "errors": 48040, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1230080, "errors": 0, "repairs_vs_native": 48040, "regressions_vs_native": 0, "regression_free_recovery": true, "perfect_recovery": true }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1230080, "errors": 0, "repairs_vs_native": 48040, "regressions_vs_native": 0, "regression_free_recovery": true, "perfect_recovery": true }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1230080, "errors": 0, "repairs_vs_native": 48040, "regressions_vs_native": 0, "regression_free_recovery": true, "perfect_recovery": true } } } }, "length32_no_projection": { "trials": [ { "seed": 0, "arm": "native", "cases": 246016, "latent_mse": 0.13263030350208282, "learned": { "correct": 107201, "accuracy": 0.43574808142559834, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 168411, "accuracy": 0.6845530372008325, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.17458903789520264, "learned": { "correct": 188890, "accuracy": 0.7677955905306972, "repairs_vs_native": 93986, "regressions_vs_native": 12297 }, "reference": { "correct": 246012, "accuracy": 0.9999837408949012, "repairs_vs_native": 77601, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 246016, "latent_mse": 0.12530693411827087, "learned": { "correct": 201659, "accuracy": 0.8196987187825182, "repairs_vs_native": 103491, "regressions_vs_native": 9033 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 77605, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 246016, "latent_mse": 0.12522199749946594, "learned": { "correct": 201564, "accuracy": 0.8193125650364204, "repairs_vs_native": 103458, "regressions_vs_native": 9095 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 77605, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 246016, "latent_mse": 0.11262127757072449, "learned": { "correct": 107853, "accuracy": 0.43839831555671177, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 191327, "accuracy": 0.7777014503121749, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.10300248116254807, "learned": { "correct": 189027, "accuracy": 0.768352464880333, "repairs_vs_native": 93421, "regressions_vs_native": 12247 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 54689, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 246016, "latent_mse": 0.074556365609169, "learned": { "correct": 199037, "accuracy": 0.8090408753902185, "repairs_vs_native": 100726, "regressions_vs_native": 9542 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 54689, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 246016, "latent_mse": 0.07449109107255936, "learned": { "correct": 199228, "accuracy": 0.8098172476586889, "repairs_vs_native": 100920, "regressions_vs_native": 9545 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 54689, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 246016, "latent_mse": 0.12941986322402954, "learned": { "correct": 131167, "accuracy": 0.5331645096253902, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 180582, "accuracy": 0.7340254292403746, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.18101570010185242, "learned": { "correct": 174706, "accuracy": 0.7101408038501561, "repairs_vs_native": 66185, "regressions_vs_native": 22646 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 65434, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 246016, "latent_mse": 0.12646415829658508, "learned": { "correct": 193777, "accuracy": 0.7876601521852237, "repairs_vs_native": 77485, "regressions_vs_native": 14875 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 65434, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 246016, "latent_mse": 0.12634246051311493, "learned": { "correct": 193292, "accuracy": 0.7856887356919875, "repairs_vs_native": 77338, "regressions_vs_native": 15213 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 65434, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 246016, "latent_mse": 0.12586075067520142, "learned": { "correct": 121188, "accuracy": 0.4926021071800208, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 172483, "accuracy": 0.7011048061914672, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.18225237727165222, "learned": { "correct": 168407, "accuracy": 0.6845367780957337, "repairs_vs_native": 66924, "regressions_vs_native": 19705 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 73533, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 246016, "latent_mse": 0.12989670038223267, "learned": { "correct": 181181, "accuracy": 0.7364602302289281, "repairs_vs_native": 74868, "regressions_vs_native": 14875 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 73533, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 246016, "latent_mse": 0.12975004315376282, "learned": { "correct": 181141, "accuracy": 0.7362976391779397, "repairs_vs_native": 74997, "regressions_vs_native": 15044 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 73533, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 246016, "latent_mse": 0.1298651546239853, "learned": { "correct": 99715, "accuracy": 0.40531916623309056, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 149048, "accuracy": 0.6058467741935484, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 246016, "latent_mse": 0.13372614979743958, "learned": { "correct": 173762, "accuracy": 0.7063036550468262, "repairs_vs_native": 86746, "regressions_vs_native": 12699 }, "reference": { "correct": 245982, "accuracy": 0.9998617976066597, "repairs_vs_native": 96934, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 246016, "latent_mse": 0.0961855798959732, "learned": { "correct": 186523, "accuracy": 0.7581742650884495, "repairs_vs_native": 96471, "regressions_vs_native": 9663 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 96968, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 246016, "latent_mse": 0.09593392163515091, "learned": { "correct": 184715, "accuracy": 0.7508251495837669, "repairs_vs_native": 95384, "regressions_vs_native": 10384 }, "reference": { "correct": 246016, "accuracy": 1.0, "repairs_vs_native": 96968, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.43839831555671177, "minimum_accuracy": 0.40531916623309056, "maximum_accuracy": 0.5331645096253902, "cases": 1230080, "correct": 567124, "errors": 662956, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.7101408038501561, "minimum_accuracy": 0.6845367780957337, "maximum_accuracy": 0.768352464880333, "cases": 1230080, "correct": 894792, "errors": 335288, "repairs_vs_native": 407262, "regressions_vs_native": 79594, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.7876601521852237, "minimum_accuracy": 0.7364602302289281, "maximum_accuracy": 0.8196987187825182, "cases": 1230080, "correct": 962177, "errors": 267903, "repairs_vs_native": 453041, "regressions_vs_native": 57988, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.7856887356919875, "minimum_accuracy": 0.7362976391779397, "maximum_accuracy": 0.8193125650364204, "cases": 1230080, "correct": 959940, "errors": 270140, "repairs_vs_native": 452097, "regressions_vs_native": 59281, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 0.7011048061914672, "minimum_accuracy": 0.6058467741935484, "maximum_accuracy": 0.7777014503121749, "cases": 1230080, "correct": 861851, "errors": 368229, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 0.9998617976066597, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1230042, "errors": 38, "repairs_vs_native": 368191, "regressions_vs_native": 0, "regression_free_recovery": true, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1230080, "errors": 0, "repairs_vs_native": 368229, "regressions_vs_native": 0, "regression_free_recovery": true, "perfect_recovery": true }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 1230080, "correct": 1230080, "errors": 0, "repairs_vs_native": 368229, "regressions_vs_native": 0, "regression_free_recovery": true, "perfect_recovery": true } } } }, "relations_X_inverse_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.00043994071893393993, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00043994071893393993, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00043994071893393993, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.00043994071893393993, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.00028816406847909093, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00028816406847909093, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00028816406847909093, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.00028816406847909093, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.00044023754890076816, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00044023754890076816, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00044023754890076816, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.00044023754890076816, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.00046122813364490867, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00046122813364490867, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00046122813364490867, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.00046122813364490867, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.00040678700315766037, "learned": { "correct": 1889, "accuracy": 0.9828303850156087, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00040678700315766037, "learned": { "correct": 1889, "accuracy": 0.9828303850156087, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00040678700315766037, "learned": { "correct": 1889, "accuracy": 0.9828303850156087, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.00040678700315766037, "learned": { "correct": 1889, "accuracy": 0.9828303850156087, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9828303850156087, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9438, "errors": 172, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9828303850156087, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9438, "errors": 172, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9828303850156087, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9438, "errors": 172, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9828303850156087, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9438, "errors": 172, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_X_inverse_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_inverse_X_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.00043992503196932375, "learned": { "correct": 1876, "accuracy": 0.9760665972944849, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00043992503196932375, "learned": { "correct": 1876, "accuracy": 0.9760665972944849, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00043992503196932375, "learned": { "correct": 1876, "accuracy": 0.9760665972944849, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.00043992503196932375, "learned": { "correct": 1876, "accuracy": 0.9760665972944849, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.00028807262424379587, "learned": { "correct": 1867, "accuracy": 0.9713839750260146, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00028807262424379587, "learned": { "correct": 1867, "accuracy": 0.9713839750260146, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00028807262424379587, "learned": { "correct": 1867, "accuracy": 0.9713839750260146, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.00028807262424379587, "learned": { "correct": 1867, "accuracy": 0.9713839750260146, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.00044021030771546066, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00044021030771546066, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00044021030771546066, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.00044021030771546066, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.00046114425640553236, "learned": { "correct": 1909, "accuracy": 0.9932362122788762, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00046114425640553236, "learned": { "correct": 1909, "accuracy": 0.9932362122788762, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00046114425640553236, "learned": { "correct": 1909, "accuracy": 0.9932362122788762, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.00046114425640553236, "learned": { "correct": 1909, "accuracy": 0.9932362122788762, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.00040691750473342836, "learned": { "correct": 1889, "accuracy": 0.9828303850156087, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00040691750473342836, "learned": { "correct": 1889, "accuracy": 0.9828303850156087, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00040691750473342836, "learned": { "correct": 1889, "accuracy": 0.9828303850156087, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.00040691750473342836, "learned": { "correct": 1889, "accuracy": 0.9828303850156087, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9828303850156087, "minimum_accuracy": 0.9713839750260146, "maximum_accuracy": 0.9932362122788762, "cases": 9610, "correct": 9435, "errors": 175, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9828303850156087, "minimum_accuracy": 0.9713839750260146, "maximum_accuracy": 0.9932362122788762, "cases": 9610, "correct": 9435, "errors": 175, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9828303850156087, "minimum_accuracy": 0.9713839750260146, "maximum_accuracy": 0.9932362122788762, "cases": 9610, "correct": 9435, "errors": 175, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9828303850156087, "minimum_accuracy": 0.9713839750260146, "maximum_accuracy": 0.9932362122788762, "cases": 9610, "correct": 9435, "errors": 175, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_inverse_X_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_R_inverse_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.00047438591718673706, "learned": { "correct": 1884, "accuracy": 0.9802289281997919, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00047438591718673706, "learned": { "correct": 1884, "accuracy": 0.9802289281997919, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00047438591718673706, "learned": { "correct": 1884, "accuracy": 0.9802289281997919, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.00047438591718673706, "learned": { "correct": 1884, "accuracy": 0.9802289281997919, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.00032659180578775704, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00032659180578775704, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00032659180578775704, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.00032659180578775704, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0004705087631009519, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004705087631009519, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0004705087631009519, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0004705087631009519, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0004905866226181388, "learned": { "correct": 1900, "accuracy": 0.9885535900104059, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004905866226181388, "learned": { "correct": 1900, "accuracy": 0.9885535900104059, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0004905866226181388, "learned": { "correct": 1900, "accuracy": 0.9885535900104059, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0004905866226181388, "learned": { "correct": 1900, "accuracy": 0.9885535900104059, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0003810607304330915, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0003810607304330915, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0003810607304330915, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0003810607304330915, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9885535900104059, "cases": 9610, "correct": 9435, "errors": 175, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9885535900104059, "cases": 9610, "correct": 9435, "errors": 175, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9885535900104059, "cases": 9610, "correct": 9435, "errors": 175, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9885535900104059, "cases": 9610, "correct": 9435, "errors": 175, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_R_inverse_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_inverse_R_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0004743974423035979, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004743974423035979, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0004743974423035979, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0004743974423035979, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0003265863051638007, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0003265863051638007, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0003265863051638007, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0003265863051638007, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0004705577448476106, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004705577448476106, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0004705577448476106, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0004705577448476106, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0004905496607534587, "learned": { "correct": 1903, "accuracy": 0.9901144640998959, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004905496607534587, "learned": { "correct": 1903, "accuracy": 0.9901144640998959, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0004905496607534587, "learned": { "correct": 1903, "accuracy": 0.9901144640998959, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0004905496607534587, "learned": { "correct": 1903, "accuracy": 0.9901144640998959, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0003810856433119625, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0003810856433119625, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0003810856433119625, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0003810856433119625, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9901144640998959, "cases": 9610, "correct": 9431, "errors": 179, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9901144640998959, "cases": 9610, "correct": 9431, "errors": 179, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9901144640998959, "cases": 9610, "correct": 9431, "errors": 179, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9901144640998959, "cases": 9610, "correct": 9431, "errors": 179, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_inverse_R_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_X_order_31_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.153458833694458, "learned": { "correct": 1391, "accuracy": 0.7237252861602498, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1860, "accuracy": 0.967741935483871, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.153458833694458, "learned": { "correct": 1391, "accuracy": 0.7237252861602498, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1860, "accuracy": 0.967741935483871, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.153458833694458, "learned": { "correct": 1391, "accuracy": 0.7237252861602498, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1860, "accuracy": 0.967741935483871, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.153458833694458, "learned": { "correct": 1391, "accuracy": 0.7237252861602498, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1860, "accuracy": 0.967741935483871, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.08993898332118988, "learned": { "correct": 1383, "accuracy": 0.7195629552549427, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.08993898332118988, "learned": { "correct": 1383, "accuracy": 0.7195629552549427, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.08993898332118988, "learned": { "correct": 1383, "accuracy": 0.7195629552549427, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.08993898332118988, "learned": { "correct": 1383, "accuracy": 0.7195629552549427, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.156606063246727, "learned": { "correct": 1427, "accuracy": 0.7424557752341311, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.156606063246727, "learned": { "correct": 1427, "accuracy": 0.7424557752341311, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.156606063246727, "learned": { "correct": 1427, "accuracy": 0.7424557752341311, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.156606063246727, "learned": { "correct": 1427, "accuracy": 0.7424557752341311, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.16481024026870728, "learned": { "correct": 1407, "accuracy": 0.7320499479708636, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.16481024026870728, "learned": { "correct": 1407, "accuracy": 0.7320499479708636, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.16481024026870728, "learned": { "correct": 1407, "accuracy": 0.7320499479708636, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.16481024026870728, "learned": { "correct": 1407, "accuracy": 0.7320499479708636, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.13535457849502563, "learned": { "correct": 1388, "accuracy": 0.7221644120707597, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1883, "accuracy": 0.9797086368366286, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.13535457849502563, "learned": { "correct": 1388, "accuracy": 0.7221644120707597, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1883, "accuracy": 0.9797086368366286, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.13535457849502563, "learned": { "correct": 1388, "accuracy": 0.7221644120707597, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1883, "accuracy": 0.9797086368366286, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.13535457849502563, "learned": { "correct": 1388, "accuracy": 0.7221644120707597, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1883, "accuracy": 0.9797086368366286, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.7237252861602498, "minimum_accuracy": 0.7195629552549427, "maximum_accuracy": 0.7424557752341311, "cases": 9610, "correct": 6996, "errors": 2614, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.7237252861602498, "minimum_accuracy": 0.7195629552549427, "maximum_accuracy": 0.7424557752341311, "cases": 9610, "correct": 6996, "errors": 2614, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.7237252861602498, "minimum_accuracy": 0.7195629552549427, "maximum_accuracy": 0.7424557752341311, "cases": 9610, "correct": 6996, "errors": 2614, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.7237252861602498, "minimum_accuracy": 0.7195629552549427, "maximum_accuracy": 0.7424557752341311, "cases": 9610, "correct": 6996, "errors": 2614, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 0.967741935483871, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9509, "errors": 101, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 0.967741935483871, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9509, "errors": 101, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 0.967741935483871, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9509, "errors": 101, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 0.967741935483871, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9509, "errors": 101, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_X_order_31_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_Y_order_31_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.21527686715126038, "learned": { "correct": 1459, "accuracy": 0.759105098855359, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.21527686715126038, "learned": { "correct": 1459, "accuracy": 0.759105098855359, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.21527686715126038, "learned": { "correct": 1459, "accuracy": 0.759105098855359, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.21527686715126038, "learned": { "correct": 1459, "accuracy": 0.759105098855359, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.12039124220609665, "learned": { "correct": 1432, "accuracy": 0.745057232049948, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.12039124220609665, "learned": { "correct": 1432, "accuracy": 0.745057232049948, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.12039124220609665, "learned": { "correct": 1432, "accuracy": 0.745057232049948, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.12039124220609665, "learned": { "correct": 1432, "accuracy": 0.745057232049948, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.2123289704322815, "learned": { "correct": 1351, "accuracy": 0.7029136316337149, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.2123289704322815, "learned": { "correct": 1351, "accuracy": 0.7029136316337149, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.2123289704322815, "learned": { "correct": 1351, "accuracy": 0.7029136316337149, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.2123289704322815, "learned": { "correct": 1351, "accuracy": 0.7029136316337149, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.22955875098705292, "learned": { "correct": 1354, "accuracy": 0.704474505723205, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.22955875098705292, "learned": { "correct": 1354, "accuracy": 0.704474505723205, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.22955875098705292, "learned": { "correct": 1354, "accuracy": 0.704474505723205, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.22955875098705292, "learned": { "correct": 1354, "accuracy": 0.704474505723205, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.16463245451450348, "learned": { "correct": 1355, "accuracy": 0.7049947970863684, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.16463245451450348, "learned": { "correct": 1355, "accuracy": 0.7049947970863684, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.16463245451450348, "learned": { "correct": 1355, "accuracy": 0.7049947970863684, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.16463245451450348, "learned": { "correct": 1355, "accuracy": 0.7049947970863684, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.7049947970863684, "minimum_accuracy": 0.7029136316337149, "maximum_accuracy": 0.759105098855359, "cases": 9610, "correct": 6951, "errors": 2659, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.7049947970863684, "minimum_accuracy": 0.7029136316337149, "maximum_accuracy": 0.759105098855359, "cases": 9610, "correct": 6951, "errors": 2659, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.7049947970863684, "minimum_accuracy": 0.7029136316337149, "maximum_accuracy": 0.759105098855359, "cases": 9610, "correct": 6951, "errors": 2659, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.7049947970863684, "minimum_accuracy": 0.7029136316337149, "maximum_accuracy": 0.759105098855359, "cases": 9610, "correct": 6951, "errors": 2659, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_Y_order_31_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_S_order_31_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.15934664011001587, "learned": { "correct": 1474, "accuracy": 0.7669094693028096, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1860, "accuracy": 0.967741935483871, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.15934664011001587, "learned": { "correct": 1474, "accuracy": 0.7669094693028096, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1860, "accuracy": 0.967741935483871, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.15934664011001587, "learned": { "correct": 1474, "accuracy": 0.7669094693028096, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1860, "accuracy": 0.967741935483871, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.15934664011001587, "learned": { "correct": 1474, "accuracy": 0.7669094693028096, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1860, "accuracy": 0.967741935483871, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.09628959745168686, "learned": { "correct": 1360, "accuracy": 0.7075962539021852, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.09628959745168686, "learned": { "correct": 1360, "accuracy": 0.7075962539021852, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.09628959745168686, "learned": { "correct": 1360, "accuracy": 0.7075962539021852, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.09628959745168686, "learned": { "correct": 1360, "accuracy": 0.7075962539021852, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.163092240691185, "learned": { "correct": 1407, "accuracy": 0.7320499479708636, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.163092240691185, "learned": { "correct": 1407, "accuracy": 0.7320499479708636, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.163092240691185, "learned": { "correct": 1407, "accuracy": 0.7320499479708636, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.163092240691185, "learned": { "correct": 1407, "accuracy": 0.7320499479708636, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.17116528749465942, "learned": { "correct": 1351, "accuracy": 0.7029136316337149, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.17116528749465942, "learned": { "correct": 1351, "accuracy": 0.7029136316337149, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.17116528749465942, "learned": { "correct": 1351, "accuracy": 0.7029136316337149, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.17116528749465942, "learned": { "correct": 1351, "accuracy": 0.7029136316337149, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.14170187711715698, "learned": { "correct": 1338, "accuracy": 0.696149843912591, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.14170187711715698, "learned": { "correct": 1338, "accuracy": 0.696149843912591, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.14170187711715698, "learned": { "correct": 1338, "accuracy": 0.696149843912591, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.14170187711715698, "learned": { "correct": 1338, "accuracy": 0.696149843912591, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.7075962539021852, "minimum_accuracy": 0.696149843912591, "maximum_accuracy": 0.7669094693028096, "cases": 9610, "correct": 6930, "errors": 2680, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.7075962539021852, "minimum_accuracy": 0.696149843912591, "maximum_accuracy": 0.7669094693028096, "cases": 9610, "correct": 6930, "errors": 2680, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.7075962539021852, "minimum_accuracy": 0.696149843912591, "maximum_accuracy": 0.7669094693028096, "cases": 9610, "correct": 6930, "errors": 2680, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.7075962539021852, "minimum_accuracy": 0.696149843912591, "maximum_accuracy": 0.7669094693028096, "cases": 9610, "correct": 6930, "errors": 2680, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 0.9516129032258065, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9455, "errors": 155, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 0.9516129032258065, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9455, "errors": 155, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 0.9516129032258065, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9455, "errors": 155, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 0.9516129032258065, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9455, "errors": 155, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_S_order_31_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_R_order_4_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0019295233068987727, "learned": { "correct": 1877, "accuracy": 0.9765868886576483, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0019295233068987727, "learned": { "correct": 1877, "accuracy": 0.9765868886576483, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0019295233068987727, "learned": { "correct": 1877, "accuracy": 0.9765868886576483, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0019295233068987727, "learned": { "correct": 1877, "accuracy": 0.9765868886576483, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0013008982641622424, "learned": { "correct": 1869, "accuracy": 0.9724245577523413, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0013008982641622424, "learned": { "correct": 1869, "accuracy": 0.9724245577523413, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0013008982641622424, "learned": { "correct": 1869, "accuracy": 0.9724245577523413, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0013008982641622424, "learned": { "correct": 1869, "accuracy": 0.9724245577523413, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0019349342910572886, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0019349342910572886, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0019349342910572886, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0019349342910572886, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0019435223657637835, "learned": { "correct": 1885, "accuracy": 0.9807492195629552, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0019435223657637835, "learned": { "correct": 1885, "accuracy": 0.9807492195629552, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0019435223657637835, "learned": { "correct": 1885, "accuracy": 0.9807492195629552, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0019435223657637835, "learned": { "correct": 1885, "accuracy": 0.9807492195629552, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0015424054581671953, "learned": { "correct": 1872, "accuracy": 0.9739854318418314, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0015424054581671953, "learned": { "correct": 1872, "accuracy": 0.9739854318418314, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0015424054581671953, "learned": { "correct": 1872, "accuracy": 0.9739854318418314, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0015424054581671953, "learned": { "correct": 1872, "accuracy": 0.9739854318418314, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9765868886576483, "minimum_accuracy": 0.9724245577523413, "maximum_accuracy": 0.9812695109261186, "cases": 9610, "correct": 9389, "errors": 221, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9765868886576483, "minimum_accuracy": 0.9724245577523413, "maximum_accuracy": 0.9812695109261186, "cases": 9610, "correct": 9389, "errors": 221, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9765868886576483, "minimum_accuracy": 0.9724245577523413, "maximum_accuracy": 0.9812695109261186, "cases": 9610, "correct": 9389, "errors": 221, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9765868886576483, "minimum_accuracy": 0.9724245577523413, "maximum_accuracy": 0.9812695109261186, "cases": 9610, "correct": 9389, "errors": 221, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_R_order_4_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_H_order_2_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.05236804485321045, "learned": { "correct": 1740, "accuracy": 0.9053069719042663, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00043889088556170464, "learned": { "correct": 1880, "accuracy": 0.9781477627471384, "repairs_vs_native": 157, "regressions_vs_native": 17 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 7.868133092758453e-14, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 154, "regressions_vs_native": 8 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 8.778294264933291e-33, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 154, "regressions_vs_native": 8 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.04659634456038475, "learned": { "correct": 1642, "accuracy": 0.854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00029049202566966414, "learned": { "correct": 1866, "accuracy": 0.9708636836628513, "repairs_vs_native": 237, "regressions_vs_native": 13 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 6.444735560105427e-14, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 238, "regressions_vs_native": 18 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 8.778294264933291e-33, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 238, "regressions_vs_native": 18 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.04901675507426262, "learned": { "correct": 1838, "accuracy": 0.9562955254942768, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0005177512648515403, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 70, "regressions_vs_native": 14 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 8.039232392851106e-14, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 70, "regressions_vs_native": 15 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 8.778294264933291e-33, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 70, "regressions_vs_native": 15 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.051233429461717606, "learned": { "correct": 1714, "accuracy": 0.8917793964620188, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00047295671538449824, "learned": { "correct": 1902, "accuracy": 0.9895941727367326, "repairs_vs_native": 196, "regressions_vs_native": 8 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 8.70073869722876e-14, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 199, "regressions_vs_native": 2 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 8.778294264933291e-33, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 199, "regressions_vs_native": 2 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.050206027925014496, "learned": { "correct": 1604, "accuracy": 0.8345473465140478, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00038041084189899266, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 299, "regressions_vs_native": 12 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 7.459083264244049e-14, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 302, "regressions_vs_native": 15 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 8.778294264933291e-33, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 302, "regressions_vs_native": 15 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.8917793964620188, "minimum_accuracy": 0.8345473465140478, "maximum_accuracy": 0.9562955254942768, "cases": 9610, "correct": 8538, "errors": 1072, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9708636836628513, "maximum_accuracy": 0.9895941727367326, "cases": 9610, "correct": 9433, "errors": 177, "repairs_vs_native": 959, "regressions_vs_native": 64, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 963, "regressions_vs_native": 58, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 963, "regressions_vs_native": 58, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_H_order_2_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_context_inverse_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.04860883578658104, "learned": { "correct": 1717, "accuracy": 0.8933402705515089, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0018080957233905792, "learned": { "correct": 1873, "accuracy": 0.9745057232049948, "repairs_vs_native": 163, "regressions_vs_native": 7 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00044333210098557174, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 166, "regressions_vs_native": 5 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.00044322150642983615, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 166, "regressions_vs_native": 5 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.04370811581611633, "learned": { "correct": 1639, "accuracy": 0.8527575442247659, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.001153237884864211, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 239, "regressions_vs_native": 16 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0002790675207506865, "learned": { "correct": 1867, "accuracy": 0.9713839750260146, "repairs_vs_native": 239, "regressions_vs_native": 11 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0002788541023619473, "learned": { "correct": 1867, "accuracy": 0.9713839750260146, "repairs_vs_native": 239, "regressions_vs_native": 11 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.045489657670259476, "learned": { "correct": 1835, "accuracy": 0.9547346514047866, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0019882419146597385, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 72, "regressions_vs_native": 19 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00045518146362155676, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 72, "regressions_vs_native": 13 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.00045487628085538745, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 72, "regressions_vs_native": 13 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.04723460599780083, "learned": { "correct": 1713, "accuracy": 0.8912591050988553, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0019212091574445367, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 191, "regressions_vs_native": 16 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00046344485599547625, "learned": { "correct": 1901, "accuracy": 0.9890738813735692, "repairs_vs_native": 195, "regressions_vs_native": 7 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.000462809344753623, "learned": { "correct": 1902, "accuracy": 0.9895941727367326, "repairs_vs_native": 196, "regressions_vs_native": 7 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.04680326208472252, "learned": { "correct": 1583, "accuracy": 0.823621227887617, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0015506012132391334, "learned": { "correct": 1887, "accuracy": 0.981789802289282, "repairs_vs_native": 314, "regressions_vs_native": 10 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00037946406519040465, "learned": { "correct": 1889, "accuracy": 0.9828303850156087, "repairs_vs_native": 318, "regressions_vs_native": 12 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0003785245935432613, "learned": { "correct": 1887, "accuracy": 0.981789802289282, "repairs_vs_native": 318, "regressions_vs_native": 14 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.8912591050988553, "minimum_accuracy": 0.823621227887617, "maximum_accuracy": 0.9547346514047866, "cases": 9610, "correct": 8487, "errors": 1123, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.981789802289282, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9823100936524454, "cases": 9610, "correct": 9398, "errors": 212, "repairs_vs_native": 979, "regressions_vs_native": 68, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9828303850156087, "minimum_accuracy": 0.9713839750260146, "maximum_accuracy": 0.9890738813735692, "cases": 9610, "correct": 9429, "errors": 181, "repairs_vs_native": 990, "regressions_vs_native": 48, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.981789802289282, "minimum_accuracy": 0.9713839750260146, "maximum_accuracy": 0.9895941727367326, "cases": 9610, "correct": 9428, "errors": 182, "repairs_vs_native": 991, "regressions_vs_native": 50, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_context_inverse_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1862, "accuracy": 0.9687825182101977, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1911, "accuracy": 0.9942767950052029, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9838709677419355, "minimum_accuracy": 0.9687825182101977, "maximum_accuracy": 0.9942767950052029, "cases": 9610, "correct": 9443, "errors": 167, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_projection_idempotent_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.002842947840690613, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.002842947840690613, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.002842947840690613, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.002842947840690613, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.004306325688958168, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.004306325688958168, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.004306325688958168, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.004306325688958168, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0016832168912515044, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0016832168912515044, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0016832168912515044, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0016832168912515044, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.006471444386988878, "learned": { "correct": 1798, "accuracy": 0.9354838709677419, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.006471444386988878, "learned": { "correct": 1798, "accuracy": 0.9354838709677419, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.006471444386988878, "learned": { "correct": 1798, "accuracy": 0.9354838709677419, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.006471444386988878, "learned": { "correct": 1798, "accuracy": 0.9354838709677419, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.006175126414746046, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.006175126414746046, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.006175126414746046, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.006175126414746046, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9354838709677419, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9083, "errors": 527, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9354838709677419, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9083, "errors": 527, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9354838709677419, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9083, "errors": 527, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9354838709677419, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9083, "errors": 527, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_projection_idempotent_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0018223634688183665, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0018223634688183665, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0018223634688183665, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0018223634688183665, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.001925184391438961, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.001925184391438961, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.001925184391438961, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.001925184391438961, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0011324064107611775, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0011324064107611775, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0011324064107611775, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0011324064107611775, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0022054745350033045, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0022054745350033045, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0022054745350033045, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0022054745350033045, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.002777342451736331, "learned": { "correct": 1788, "accuracy": 0.9302809573361083, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.002777342451736331, "learned": { "correct": 1788, "accuracy": 0.9302809573361083, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.002777342451736331, "learned": { "correct": 1788, "accuracy": 0.9302809573361083, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.002777342451736331, "learned": { "correct": 1788, "accuracy": 0.9302809573361083, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9516129032258065, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9197, "errors": 413, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9516129032258065, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9197, "errors": 413, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9516129032258065, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9197, "errors": 413, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9516129032258065, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9197, "errors": 413, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_rotation_conjugates_translation_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.001047979574650526, "learned": { "correct": 1879, "accuracy": 0.977627471383975, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.001047979574650526, "learned": { "correct": 1879, "accuracy": 0.977627471383975, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.001047979574650526, "learned": { "correct": 1879, "accuracy": 0.977627471383975, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.001047979574650526, "learned": { "correct": 1879, "accuracy": 0.977627471383975, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0006978387827984989, "learned": { "correct": 1870, "accuracy": 0.9729448491155047, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0006978387827984989, "learned": { "correct": 1870, "accuracy": 0.9729448491155047, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0006978387827984989, "learned": { "correct": 1870, "accuracy": 0.9729448491155047, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0006978387827984989, "learned": { "correct": 1870, "accuracy": 0.9729448491155047, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.001049215323291719, "learned": { "correct": 1892, "accuracy": 0.9843912591050988, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.001049215323291719, "learned": { "correct": 1892, "accuracy": 0.9843912591050988, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.001049215323291719, "learned": { "correct": 1892, "accuracy": 0.9843912591050988, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.001049215323291719, "learned": { "correct": 1892, "accuracy": 0.9843912591050988, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0010905435774475336, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0010905435774475336, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0010905435774475336, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0010905435774475336, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0008531027706339955, "learned": { "correct": 1882, "accuracy": 0.9791883454734651, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0008531027706339955, "learned": { "correct": 1882, "accuracy": 0.9791883454734651, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0008531027706339955, "learned": { "correct": 1882, "accuracy": 0.9791883454734651, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0008531027706339955, "learned": { "correct": 1882, "accuracy": 0.9791883454734651, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9791883454734651, "minimum_accuracy": 0.9729448491155047, "maximum_accuracy": 0.9843912591050988, "cases": 9610, "correct": 9414, "errors": 196, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9791883454734651, "minimum_accuracy": 0.9729448491155047, "maximum_accuracy": 0.9843912591050988, "cases": 9610, "correct": 9414, "errors": 196, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9791883454734651, "minimum_accuracy": 0.9729448491155047, "maximum_accuracy": 0.9843912591050988, "cases": 9610, "correct": 9414, "errors": 196, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9791883454734651, "minimum_accuracy": 0.9729448491155047, "maximum_accuracy": 0.9843912591050988, "cases": 9610, "correct": 9414, "errors": 196, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "relations_rotation_conjugates_translation_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.00014377859770320356, "learned": { "correct": 1882, "accuracy": 0.9791883454734651, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00014377859770320356, "learned": { "correct": 1882, "accuracy": 0.9791883454734651, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00014377859770320356, "learned": { "correct": 1882, "accuracy": 0.9791883454734651, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.00014377859770320356, "learned": { "correct": 1882, "accuracy": 0.9791883454734651, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 9.160277113551274e-05, "learned": { "correct": 1861, "accuracy": 0.9682622268470343, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 9.160277113551274e-05, "learned": { "correct": 1861, "accuracy": 0.9682622268470343, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 9.160277113551274e-05, "learned": { "correct": 1861, "accuracy": 0.9682622268470343, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 9.160277113551274e-05, "learned": { "correct": 1861, "accuracy": 0.9682622268470343, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.00014314896543510258, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00014314896543510258, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00014314896543510258, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.00014314896543510258, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0001532083988422528, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0001532083988422528, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0001532083988422528, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0001532083988422528, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.00012211402645334601, "learned": { "correct": 1892, "accuracy": 0.9843912591050988, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00012211402645334601, "learned": { "correct": 1892, "accuracy": 0.9843912591050988, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00012211402645334601, "learned": { "correct": 1892, "accuracy": 0.9843912591050988, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.00012211402645334601, "learned": { "correct": 1892, "accuracy": 0.9843912591050988, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9843912591050988, "minimum_accuracy": 0.9682622268470343, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9437, "errors": 173, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9843912591050988, "minimum_accuracy": 0.9682622268470343, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9437, "errors": 173, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9843912591050988, "minimum_accuracy": 0.9682622268470343, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9437, "errors": 173, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9843912591050988, "minimum_accuracy": 0.9682622268470343, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9437, "errors": 173, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "noncommuting_translation_rotation_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.00045167922507971525, "learned": { "correct": 1880, "accuracy": 0.9781477627471384, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00045167922507971525, "learned": { "correct": 1880, "accuracy": 0.9781477627471384, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00045167922507971525, "learned": { "correct": 1880, "accuracy": 0.9781477627471384, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.00045167922507971525, "learned": { "correct": 1880, "accuracy": 0.9781477627471384, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0002932428615167737, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0002932428615167737, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0002932428615167737, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0002932428615167737, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0004568008880596608, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004568008880596608, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0004568008880596608, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0004568008880596608, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.000465080258436501, "learned": { "correct": 1902, "accuracy": 0.9895941727367326, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.000465080258436501, "learned": { "correct": 1902, "accuracy": 0.9895941727367326, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.000465080258436501, "learned": { "correct": 1902, "accuracy": 0.9895941727367326, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.000465080258436501, "learned": { "correct": 1902, "accuracy": 0.9895941727367326, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0003656866028904915, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0003656866028904915, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0003656866028904915, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0003656866028904915, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9895941727367326, "cases": 9610, "correct": 9432, "errors": 178, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9895941727367326, "cases": 9610, "correct": 9432, "errors": 178, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9895941727367326, "cases": 9610, "correct": 9432, "errors": 178, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9895941727367326, "cases": 9610, "correct": 9432, "errors": 178, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "noncommuting_translation_rotation_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0004629493923857808, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004629493923857808, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0004629493923857808, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0004629493923857808, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.00031465940992347896, "learned": { "correct": 1867, "accuracy": 0.9713839750260146, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00031465940992347896, "learned": { "correct": 1867, "accuracy": 0.9713839750260146, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00031465940992347896, "learned": { "correct": 1867, "accuracy": 0.9713839750260146, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.00031465940992347896, "learned": { "correct": 1867, "accuracy": 0.9713839750260146, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0004646585148293525, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004646585148293525, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0004646585148293525, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0004646585148293525, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.000483861134853214, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.000483861134853214, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.000483861134853214, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.000483861134853214, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0004079454520251602, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004079454520251602, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0004079454520251602, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0004079454520251602, "learned": { "correct": 1888, "accuracy": 0.9823100936524454, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9713839750260146, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9436, "errors": 174, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9713839750260146, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9436, "errors": 174, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9713839750260146, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9436, "errors": 174, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9823100936524454, "minimum_accuracy": 0.9713839750260146, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9436, "errors": 174, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "noncommuting_context_flip_translation_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.01576877199113369, "learned": { "correct": 1836, "accuracy": 0.9552549427679501, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00044814354623667896, "learned": { "correct": 1879, "accuracy": 0.977627471383975, "repairs_vs_native": 55, "regressions_vs_native": 12 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00011195740080438554, "learned": { "correct": 1884, "accuracy": 0.9802289281997919, "repairs_vs_native": 58, "regressions_vs_native": 10 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.00011195738625247031, "learned": { "correct": 1884, "accuracy": 0.9802289281997919, "repairs_vs_native": 58, "regressions_vs_native": 10 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.01383622270077467, "learned": { "correct": 1846, "accuracy": 0.9604578563995838, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00030604362837038934, "learned": { "correct": 1868, "accuracy": 0.9719042663891779, "repairs_vs_native": 43, "regressions_vs_native": 21 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 7.514678145525977e-05, "learned": { "correct": 1861, "accuracy": 0.9682622268470343, "repairs_vs_native": 38, "regressions_vs_native": 23 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 7.514641038142145e-05, "learned": { "correct": 1861, "accuracy": 0.9682622268470343, "repairs_vs_native": 38, "regressions_vs_native": 23 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.014485816471278667, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004915745812468231, "learned": { "correct": 1894, "accuracy": 0.9854318418314256, "repairs_vs_native": 15, "regressions_vs_native": 12 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00011449812882347032, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 16, "regressions_vs_native": 12 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.00011449787416495383, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 16, "regressions_vs_native": 12 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.014938103035092354, "learned": { "correct": 1848, "accuracy": 0.9614984391259105, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004907205584459007, "learned": { "correct": 1908, "accuracy": 0.9927159209157128, "repairs_vs_native": 65, "regressions_vs_native": 5 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00012068822252331302, "learned": { "correct": 1909, "accuracy": 0.9932362122788762, "repairs_vs_native": 66, "regressions_vs_native": 5 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0001206871384056285, "learned": { "correct": 1909, "accuracy": 0.9932362122788762, "repairs_vs_native": 66, "regressions_vs_native": 5 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.01508862990885973, "learned": { "correct": 1815, "accuracy": 0.9443288241415192, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004113180621061474, "learned": { "correct": 1892, "accuracy": 0.9843912591050988, "repairs_vs_native": 86, "regressions_vs_native": 9 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00010257347457809374, "learned": { "correct": 1892, "accuracy": 0.9843912591050988, "repairs_vs_native": 86, "regressions_vs_native": 9 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0001025732490234077, "learned": { "correct": 1892, "accuracy": 0.9843912591050988, "repairs_vs_native": 86, "regressions_vs_native": 9 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9604578563995838, "minimum_accuracy": 0.9443288241415192, "maximum_accuracy": 0.9838709677419355, "cases": 9610, "correct": 9236, "errors": 374, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9843912591050988, "minimum_accuracy": 0.9719042663891779, "maximum_accuracy": 0.9927159209157128, "cases": 9610, "correct": 9441, "errors": 169, "repairs_vs_native": 264, "regressions_vs_native": 59, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9843912591050988, "minimum_accuracy": 0.9682622268470343, "maximum_accuracy": 0.9932362122788762, "cases": 9610, "correct": 9441, "errors": 169, "repairs_vs_native": 264, "regressions_vs_native": 59, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9843912591050988, "minimum_accuracy": 0.9682622268470343, "maximum_accuracy": 0.9932362122788762, "cases": 9610, "correct": 9441, "errors": 169, "repairs_vs_native": 264, "regressions_vs_native": 59, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "noncommuting_context_flip_translation_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.01575334556400776, "learned": { "correct": 1874, "accuracy": 0.9750260145681582, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00044234489905647933, "learned": { "correct": 1878, "accuracy": 0.9771071800208116, "repairs_vs_native": 15, "regressions_vs_native": 11 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.000111952300358098, "learned": { "correct": 1882, "accuracy": 0.9791883454734651, "repairs_vs_native": 18, "regressions_vs_native": 10 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.00011182307207491249, "learned": { "correct": 1882, "accuracy": 0.9791883454734651, "repairs_vs_native": 18, "regressions_vs_native": 10 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.01384420320391655, "learned": { "correct": 1846, "accuracy": 0.9604578563995838, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00028617511270567775, "learned": { "correct": 1866, "accuracy": 0.9708636836628513, "repairs_vs_native": 33, "regressions_vs_native": 13 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 7.516866753576323e-05, "learned": { "correct": 1863, "accuracy": 0.9693028095733611, "repairs_vs_native": 33, "regressions_vs_native": 16 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 7.503391680074856e-05, "learned": { "correct": 1863, "accuracy": 0.9693028095733611, "repairs_vs_native": 33, "regressions_vs_native": 16 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.014504632912576199, "learned": { "correct": 1899, "accuracy": 0.9880332986472424, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00048698176397010684, "learned": { "correct": 1893, "accuracy": 0.9849115504682622, "repairs_vs_native": 8, "regressions_vs_native": 14 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00011450953024905175, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 8, "regressions_vs_native": 12 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.00011430623271735385, "learned": { "correct": 1895, "accuracy": 0.985952133194589, "repairs_vs_native": 8, "regressions_vs_native": 12 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.014923559501767159, "learned": { "correct": 1856, "accuracy": 0.9656607700312175, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0004684363375417888, "learned": { "correct": 1902, "accuracy": 0.9895941727367326, "repairs_vs_native": 55, "regressions_vs_native": 9 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00012069241347489879, "learned": { "correct": 1906, "accuracy": 0.991675338189386, "repairs_vs_native": 57, "regressions_vs_native": 7 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.00012037535634590313, "learned": { "correct": 1906, "accuracy": 0.991675338189386, "repairs_vs_native": 57, "regressions_vs_native": 7 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.015012136660516262, "learned": { "correct": 1837, "accuracy": 0.9557752341311134, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0003820499696303159, "learned": { "correct": 1886, "accuracy": 0.9812695109261186, "repairs_vs_native": 64, "regressions_vs_native": 15 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00010254752851324156, "learned": { "correct": 1885, "accuracy": 0.9807492195629552, "repairs_vs_native": 65, "regressions_vs_native": 17 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0001020834970404394, "learned": { "correct": 1885, "accuracy": 0.9807492195629552, "repairs_vs_native": 65, "regressions_vs_native": 17 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9656607700312175, "minimum_accuracy": 0.9557752341311134, "maximum_accuracy": 0.9880332986472424, "cases": 9610, "correct": 9312, "errors": 298, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9812695109261186, "minimum_accuracy": 0.9708636836628513, "maximum_accuracy": 0.9895941727367326, "cases": 9610, "correct": 9425, "errors": 185, "repairs_vs_native": 175, "regressions_vs_native": 62, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9807492195629552, "minimum_accuracy": 0.9693028095733611, "maximum_accuracy": 0.991675338189386, "cases": 9610, "correct": 9431, "errors": 179, "repairs_vs_native": 181, "regressions_vs_native": 62, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9807492195629552, "minimum_accuracy": 0.9693028095733611, "maximum_accuracy": 0.991675338189386, "cases": 9610, "correct": 9431, "errors": 179, "repairs_vs_native": 181, "regressions_vs_native": 62, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "noncommuting_projection_rotation_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.001909842947497964, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.001909842947497964, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.001909842947497964, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.001909842947497964, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.001932003302499652, "learned": { "correct": 1612, "accuracy": 0.8387096774193549, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.001932003302499652, "learned": { "correct": 1612, "accuracy": 0.8387096774193549, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.001932003302499652, "learned": { "correct": 1612, "accuracy": 0.8387096774193549, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.001932003302499652, "learned": { "correct": 1612, "accuracy": 0.8387096774193549, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0012721478706225753, "learned": { "correct": 1364, "accuracy": 0.7096774193548387, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0012721478706225753, "learned": { "correct": 1364, "accuracy": 0.7096774193548387, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0012721478706225753, "learned": { "correct": 1364, "accuracy": 0.7096774193548387, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0012721478706225753, "learned": { "correct": 1364, "accuracy": 0.7096774193548387, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.00233064079657197, "learned": { "correct": 1787, "accuracy": 0.9297606659729448, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.00233064079657197, "learned": { "correct": 1787, "accuracy": 0.9297606659729448, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.00233064079657197, "learned": { "correct": 1787, "accuracy": 0.9297606659729448, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.00233064079657197, "learned": { "correct": 1787, "accuracy": 0.9297606659729448, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0033124578185379505, "learned": { "correct": 1798, "accuracy": 0.9354838709677419, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0033124578185379505, "learned": { "correct": 1798, "accuracy": 0.9354838709677419, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0033124578185379505, "learned": { "correct": 1798, "accuracy": 0.9354838709677419, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0033124578185379505, "learned": { "correct": 1798, "accuracy": 0.9354838709677419, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9297606659729448, "minimum_accuracy": 0.7096774193548387, "maximum_accuracy": 1.0, "cases": 9610, "correct": 8483, "errors": 1127, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9297606659729448, "minimum_accuracy": 0.7096774193548387, "maximum_accuracy": 1.0, "cases": 9610, "correct": 8483, "errors": 1127, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9297606659729448, "minimum_accuracy": 0.7096774193548387, "maximum_accuracy": 1.0, "cases": 9610, "correct": 8483, "errors": 1127, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9297606659729448, "minimum_accuracy": 0.7096774193548387, "maximum_accuracy": 1.0, "cases": 9610, "correct": 8483, "errors": 1127, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "noncommuting_projection_rotation_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0020633749663829803, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0020633749663829803, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0020633749663829803, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0020633749663829803, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.002123153768479824, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.002123153768479824, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.002123153768479824, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.002123153768479824, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0014053619233891368, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0014053619233891368, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0014053619233891368, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0014053619233891368, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0024144414346665144, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0024144414346665144, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0024144414346665144, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0024144414346665144, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.002981959143653512, "learned": { "correct": 1774, "accuracy": 0.922996878251821, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.002981959143653512, "learned": { "correct": 1774, "accuracy": 0.922996878251821, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.002981959143653512, "learned": { "correct": 1774, "accuracy": 0.922996878251821, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.002981959143653512, "learned": { "correct": 1774, "accuracy": 0.922996878251821, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9516129032258065, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9183, "errors": 427, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9516129032258065, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9183, "errors": 427, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9516129032258065, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9183, "errors": 427, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9516129032258065, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9183, "errors": 427, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "noncommuting_projection_translation_left": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0020709731616079807, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0020709731616079807, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0020709731616079807, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0020709731616079807, "learned": { "correct": 1767, "accuracy": 0.9193548387096774, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0021170596592128277, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0021170596592128277, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0021170596592128277, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0021170596592128277, "learned": { "correct": 1829, "accuracy": 0.9516129032258065, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0014163098530843854, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0014163098530843854, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0014163098530843854, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0014163098530843854, "learned": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.0024110444355756044, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0024110444355756044, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0024110444355756044, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.0024110444355756044, "learned": { "correct": 1891, "accuracy": 0.9838709677419355, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.002988053485751152, "learned": { "correct": 1780, "accuracy": 0.9261186264308012, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.002988053485751152, "learned": { "correct": 1780, "accuracy": 0.9261186264308012, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.002988053485751152, "learned": { "correct": 1780, "accuracy": 0.9261186264308012, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.002988053485751152, "learned": { "correct": 1780, "accuracy": 0.9261186264308012, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.9516129032258065, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9189, "errors": 421, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.9516129032258065, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9189, "errors": 421, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.9516129032258065, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9189, "errors": 421, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.9516129032258065, "minimum_accuracy": 0.9193548387096774, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9189, "errors": 421, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } }, "noncommuting_projection_translation_right": { "trials": [ { "seed": 0, "arm": "native", "cases": 1922, "latent_mse": 0.0017528418684378266, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0017528418684378266, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0017528418684378266, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 0, "arm": "refit", "cases": 1922, "latent_mse": 0.0017528418684378266, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "native", "cases": 1922, "latent_mse": 0.0018535119015723467, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0018535119015723467, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0018535119015723467, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 1, "arm": "refit", "cases": 1922, "latent_mse": 0.0018535119015723467, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "native", "cases": 1922, "latent_mse": 0.0012385575100779533, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0012385575100779533, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0012385575100779533, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 2, "arm": "refit", "cases": 1922, "latent_mse": 0.0012385575100779533, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "native", "cases": 1922, "latent_mse": 0.002174913650378585, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.002174913650378585, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "latent_only", "cases": 1922, "latent_mse": 0.002174913650378585, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 3, "arm": "refit", "cases": 1922, "latent_mse": 0.002174913650378585, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "native", "cases": 1922, "latent_mse": 0.0029178946278989315, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "prediction_latent", "cases": 1922, "latent_mse": 0.0029178946278989315, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "latent_only", "cases": 1922, "latent_mse": 0.0029178946278989315, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } }, { "seed": 4, "arm": "refit", "cases": 1922, "latent_mse": 0.0029178946278989315, "learned": { "correct": 0, "accuracy": 0.0, "repairs_vs_native": 0, "regressions_vs_native": 0 }, "reference": { "correct": 1922, "accuracy": 1.0, "repairs_vs_native": 0, "regressions_vs_native": 0 } } ], "aggregate": { "learned": { "native": { "median_accuracy": 0.0, "minimum_accuracy": 0.0, "maximum_accuracy": 0.0, "cases": 9610, "correct": 0, "errors": 9610, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 0.0, "minimum_accuracy": 0.0, "maximum_accuracy": 0.0, "cases": 9610, "correct": 0, "errors": 9610, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 0.0, "minimum_accuracy": 0.0, "maximum_accuracy": 0.0, "cases": 9610, "correct": 0, "errors": 9610, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 0.0, "minimum_accuracy": 0.0, "maximum_accuracy": 0.0, "cases": 9610, "correct": 0, "errors": 9610, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } }, "reference": { "native": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "prediction_latent": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "latent_only": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false }, "refit": { "median_accuracy": 1.0, "minimum_accuracy": 1.0, "maximum_accuracy": 1.0, "cases": 9610, "correct": 9610, "errors": 0, "repairs_vs_native": 0, "regressions_vs_native": 0, "regression_free_recovery": false, "perfect_recovery": false } } } } }, "retained_artifact_hashes": { "final-models.npz": "8a584daa12513ef57de90e3e2d9f67ac1ed899597374492ef61ec3c54f3be319", "base-final.npz": "f1fd4f435175db39d729c868271a6006b0bb613a2f0d82bee20dcef612cf32fa", "initial.npz": "505c354d1559899731e94b8b164ed0e347c4d60c2f3083d8863fa249ae4af4b3", "training.json": "e95879de360db64b0976f5c3e71808b79cd8306456c71ba29c8a76307b7deb4c", "evaluation-summary.json": "276b91c60badaa891e12b835cbb62aa44317053462ed85a08c0ecc91070ceb92", "training-cuda-trace.json": "090bdabbeb59cf9445d6bcede157427cda3190b89e96da3efd55bfe4c1a5a29e", "device-activity.json": "faf9df405e0c86f4a7cbc56830377e0201f43d06a10bae4d9834799a67b7db08", "parity.json": "fe824796e66e8f5b4765cfe37473412764d00e6d291c78b2671d259ec9cc3459" }, "limitations": [ "Supplied coordinates in one finite operator family; not language understanding", "Device activity samples are device-wide; no GPU speedup baseline was measured", "Independent replay checks every word at 16 fixed initial states, on the same host", "Reference and learned decoders have different outcomes; projection failures remain", "Neither topology nor diffusion, NPU, learned coordinates or another harness was tested" ] }