Delay-Gain Certified Recurrent Block / report_bench_2026-09-02T010836.md

Mechanism confirmed, baseline not beaten

Raw ⬇ ZIP

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

(универсальный)

Ответ агента:

{ "worked": false, "confidence": 9, "verdict": "The Jacobian dissipativity penalty reduced trained-model local perturbation gain by 3.9%, confirming the intended mechanism. It significantly worsened test MSE on the matched dynamics benchmark, so this idea did not win.", "metrics": { "baseline": "Best sweep lr=0.005, weight_decay=0.0001; mean test MSE 0.000662281.", "idea": "Best idea lr=0.005; mean test MSE 0.000777429; paired delta +0.000115149, p=0.0081, 0/8 wins; local gain 0.405636 versus baseline 0.422036, confirmed=true." }, "how_to_run": "/home/maxwelhelp/main/bin/python3 bench_experiment.py", "files": [ "bench_experiment.py", "bench_report.json" ], "limitations": "Only the built-in pendulum dynamics track and rnn_small GRU were tested. The implementation is a local discrete Jacobian-gain proxy rather than a full delay-dependent Lyapunov-Krasovskii SDP; explicit delayed buffers, independent delays, global Jacobian bounds, and broader model/task sweeps were not tested.", "bench_report": { "bench_version": 1, "track": "dynamics", "model": "rnn_small", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.005, "weight_decay": 0.0001 }, "mean": 0.0006622805813094601 }, "idea": { "best_lr": 0.005, "mean": 0.0007774294572300278 }, "comparison": { "delta_mean": 0.00011514887592056766, "p_value": 0.0081, "idea_wins": 0, "n_pairs": 8, "verdict": "idea worse (significant)", "system_worked": false }, "mechanism_signature": { "baseline_mean_local_gain": 0.42203639075160027, "idea_mean_local_gain": 0.4056360647082329, "relative_reduction": 0.0388599808044046, "confirmed": true } }, "system_verdict": "partial", "practical_verdict": "harms", "mechanism_ok": 1, "system_judged": true }