{ "bench_version": 1, "track": "dynamics", "model": "rnn_small", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.01 }, "sweep": [ { "cfg": { "lr": 0.001 }, "mean": 0.5504409596323967 }, { "cfg": { "lr": 0.003 }, "mean": 0.2672337554395199 }, { "cfg": { "lr": 0.01 }, "mean": 0.0029119684331817552 } ], "full": { "mean": 0.0029119684331817552, "std": 0.0019628704033910396, "per_seed": [ 0.0013499484630301595, 0.002197778085246682, 0.0036032795906066895, 0.00771956006065011, 0.0013576406054198742, 0.0016320815775543451, 0.002993914531543851, 0.0024415445514023304 ], "n": 8 } }, "idea": { "mean": 0.2629458736628294, "std": 0.06242340567924847, "per_seed": [ 0.3094339966773987, 0.3548787236213684, 0.20706547796726227, 0.3201036751270294, 0.2438417375087738, 0.22770699858665466, 0.15347930788993835, 0.2870570719242096 ], "n": 8, "selected_cfg": { "lr": 0.01 }, "trials": [ { "cfg": { "lr": 0.001 }, "mean": 0.6568634137511253, "std": 0.09002834001247538, "per_seed": [ 0.5851488709449768, 0.6946209669113159, 0.6762194633483887, 0.8284632563591003, 0.5672825574874878, 0.6657136678695679, 0.525349497795105, 0.7121090292930603 ] }, { "cfg": { "lr": 0.003 }, "mean": 0.527643658220768, "std": 0.12691734560746284, "per_seed": [ 0.49636369943618774, 0.5941106081008911, 0.561820387840271, 0.7420464754104614, 0.4776306748390198, 0.4868730902671814, 0.26752692461013794, 0.5947774052619934 ] }, { "cfg": { "lr": 0.01 }, "mean": 0.2629458736628294, "std": 0.06242340567924847, "per_seed": [ 0.3094339966773987, 0.3548787236213684, 0.20706547796726227, 0.3201036751270294, 0.2438417375087738, 0.22770699858665466, 0.15347930788993835, 0.2870570719242096 ] } ] }, "comparison": { "delta_mean": 0.26003390522964764, "idea_wins": 0, "n_pairs": 8, "per_seed_diffs": [ 0.3080840482143685, 0.3526809455361217, 0.20346219837665558, 0.3123841150663793, 0.24248409690335393, 0.22607491700910032, 0.1504853933583945, 0.28461552737280726 ], "p_value": 0.0081, "mde": 0.05555371377308973, "mde_rel_pct": 1907.771840520575, "verdict": "idea worse (significant)", "system_worked": false }, "mechanism_signature": { "prediction": "continuation should reduce late-window loss oscillation while remaining on stable side", "baseline_feature_mean": 0.583179637221284, "idea_feature_mean": 0.16445192880727005, "relative_feature_reduction": 0.7180081088023521, "observed_final_lr_mean": 0.00029296875000000004, "n_trained_models": 16, "confirmed": true }, "bench_report": { "track_match": "stability/control -> dynamics", "paired_seeds": [ 0, 1, 2, 3, 4, 5, 6, 7 ], "baseline_grid": [ { "lr": 0.001 }, { "lr": 0.003 }, { "lr": 0.01 } ], "idea_grid": [ { "lr": 0.001 }, { "lr": 0.003 }, { "lr": 0.01 } ], "weight_decay": 0.0001, "epochs": 18, "batch_size": 64, "signature_rows": [ { "seed": 0, "baseline_F_osc": 0.6125851979328734, "idea_F_osc": 0.13701609105038687, "idea_final_lr": 0.000625, "baseline_test_mse": 0.0013499484630301595, "idea_test_mse": 0.3094339966773987 }, { "seed": 1, "baseline_F_osc": 0.6589754319853819, "idea_F_osc": 0.148168541527418, "idea_final_lr": 7.8125e-05, "baseline_test_mse": 0.002197778085246682, "idea_test_mse": 0.3548787236213684 }, { "seed": 2, "baseline_F_osc": 0.4906425914915999, "idea_F_osc": 0.15470220077295294, "idea_final_lr": 0.0003125, "baseline_test_mse": 0.0036032795906066895, "idea_test_mse": 0.20706547796726227 }, { "seed": 3, "baseline_F_osc": 0.5542223686616028, "idea_F_osc": 0.1682112135875921, "idea_final_lr": 3.90625e-05, "baseline_test_mse": 0.00771956006065011, "idea_test_mse": 0.3201036751270294 }, { "seed": 4, "baseline_F_osc": 0.5676767941860419, "idea_F_osc": 0.17183414086022114, "idea_final_lr": 1.953125e-05, "baseline_test_mse": 0.0013576406054198742, "idea_test_mse": 0.2438417375087738 }, { "seed": 5, "baseline_F_osc": 0.6356397616859335, "idea_F_osc": 0.14964675665547397, "idea_final_lr": 0.000625, "baseline_test_mse": 0.0016320815775543451, "idea_test_mse": 0.22770699858665466 }, { "seed": 6, "baseline_F_osc": 0.480638587229412, "idea_F_osc": 0.18274592980114265, "idea_final_lr": 0.000625, "baseline_test_mse": 0.002993914531543851, "idea_test_mse": 0.15347930788993835 }, { "seed": 7, "baseline_F_osc": 0.6650563645974263, "idea_F_osc": 0.20329055620297268, "idea_final_lr": 1.953125e-05, "baseline_test_mse": 0.0024415445514023304, "idea_test_mse": 0.2870570719242096 } ] } }