Lyapunov Fading-Memory Optimizer / report_bench_2026-09-03T131319.md

Failed on benchmark

Raw ⬇ ZIP

Стенд-проверка (stage-2) · промт оператора:

(универсальный)

Ответ агента:

{ "worked": false, "confidence": 9, "verdict": "Implemented the Lyapunov fading-memory optimizer as a matched intervention on the registered actuated-pendulum dynamics track using the shared rnn_small architecture. Across 8 paired seeds, it was significantly worse than tuned momentum SGD: mean test MSE 0.008838 versus 0.002435, paired delta +0.0064025, permutation p=0.037. The mechanism signature was also not confirmed because the observed trained-model memory-force response increased rather than following the predicted exponential decay.", "metrics": { "baseline": "Tuned SGD with momentum=0.9 and lr=0.006; mean test MSE 0.0024354744236916304, std 0.0014542907868172839.", "idea": "Fading memory with lr=0.003, momentum=0.9, kappa=0.5, beta=2.0; mean test MSE 0.008837972913170233, std 0.006216926394780772.", "comparison": "delta_mean=+0.006402498489478603; idea_wins=1/8; permutation p_value=0.037; verdict=idea worse (significant)." }, "how_to_run": "/home/maxwelhelp/main/bin/python3 stage2_dynamics.py", "files": [ "stage2_dynamics.py", "bench_report.json" ], "limitations": "Only the registered built-in dynamics track was evaluated; tabular, vision, and sequence transfer were not tested. The mechanism probe used perturbation responses of trained models rather than direct optimizer trajectory logging. The memory settings were evaluated on a small fixed grid and no broader optimizer-family comparison was run.", "bench_report": { "bench_version": 1, "track": "dynamics", "model": "rnn_small", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.006, "momentum": 0.9 }, "sweep": [ { "cfg": { "lr": 0.001, "momentum": 0.0 }, "mean": 0.6811865866184235 }, { "cfg": { "lr": 0.001, "momentum": 0.9 }, "mean": 0.2693689875304699 }, { "cfg": { "lr": 0.003, "momentum": 0.0 }, "mean": 0.5496408864855766 }, { "cfg": { "lr": 0.003, "momentum": 0.9 }, "mean": 0.005768774193711579 }, { "cfg": { "lr": 0.006, "momentum": 0.0 }, "mean": 0.3778964877128601 }, { "cfg": { "lr": 0.006, "momentum": 0.9 }, "mean": 0.0021511380036827177 } ], "full": { "mean": 0.0024354744236916304, "std": 0.0014542907868172839, "per_seed": [ 0.004145679529756308, 0.002980618504807353, 0.0008099595434032381, 0.000668294436763972, 0.0030060403514653444, 0.003159687388688326, 0.004236925859004259, 0.00047658977564424276 ], "n": 8 } }, "idea": { "mean": 0.008837972913170233, "std": 0.006216926394780772, "per_seed": [ 0.0022076719906181097, 0.006650018040090799, 0.011529006995260715, 0.02273235283792019, 0.0036975364200770855, 0.006488712504506111, 0.004945824388414621, 0.012452660128474236 ], "n": 8, "best_cfg": { "lr": 0.003, "momentum": 0.9, "kappa": 0.5, "beta": 2.0 } }, "comparison": { "delta_mean": 0.006402498489478603, "idea_wins": 1, "n_pairs": 8, "per_seed_diffs": [ -0.001938007539138198, 0.0036693995352834463, 0.010719047451857477, 0.022064058401156217, 0.0006914960686117411, 0.0033290251158177853, 0.0007088985294103622, 0.011976070352829993 ], "p_value": 0.037, "mde": 0.006677981109294924, "mde_rel_pct": 274.19631445657353, "verdict": "idea worse (significant)", "system_worked": false }, "track_choice": "Lyapunov/stability optimizer structurally matches actuated pendulum dynamics track", "mechanism_signature": { "prediction": "exponential memory state decay rate beta*lr in discrete small-step response", "predicted_decay_rate": 0.006, "observed_decay_rate_mean": -0.10988588097760517, "observed_force_ratio_final_mean": 3.349307292855885, "observed_force_ratio_final_per_seed": [ 3.3285459078945556, 3.357568656130888, 3.3451523205798117, 3.365962286818286 ], "confirmed": false } }, "system_verdict": "failed", "practical_verdict": "harms", "mechanism_ok": 0, "system_judged": true }