{ "bench_version": 1, "track": "tabular", "model": "mlp_tiny", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.01 }, "sweep": [ { "cfg": { "lr": 0.001 }, "mean": 18.263182520866394 }, { "cfg": { "lr": 0.003 }, "mean": 13.55293893814087 }, { "cfg": { "lr": 0.01 }, "mean": 7.931941092014313 } ], "full": { "mean": 7.931941092014313, "std": 0.8101161551333251, "per_seed": [ 7.2100510597229, 8.814193725585938, 7.362153053283691, 7.630332946777344, 6.520580291748047, 8.472823143005371, 8.71557903289795, 8.729815483093262 ], "n": 8 } }, "idea": { "per_seed": [ 122.21617126464844, 161.85421752929688, 105.60208129882812, 147.32037353515625, 176.8717498779297, 125.64606475830078, 179.69190979003906, 218.0858612060547 ], "mean": 154.66105365753174, "best_cfg": { "lr": 0.003 }, "sweep": [ { "cfg": { "lr": 0.001 }, "mean": 162.4086561203003, "per_seed": [ 120.88668060302734, 152.7420654296875, 81.68341064453125, 138.7200927734375, 179.13693237304688, 229.2186279296875, 181.1822967529297, 215.6991424560547 ] }, { "cfg": { "lr": 0.003 }, "mean": 154.66105365753174, "per_seed": [ 122.21617126464844, 161.85421752929688, 105.60208129882812, 147.32037353515625, 176.8717498779297, 125.64606475830078, 179.69190979003906, 218.0858612060547 ] }, { "cfg": { "lr": 0.01 }, "mean": 159.42518997192383, "per_seed": [ 191.97760009765625, 190.82362365722656, 185.17913818359375, 175.83287048339844, 172.7498016357422, 86.1894302368164, 177.15773010253906, 95.49132537841797 ] } ] }, "comparison": { "delta_mean": 146.72911256551743, "idea_wins": 0, "n_pairs": 8, "per_seed_diffs": [ 115.00612020492554, 153.04002380371094, 98.23992824554443, 139.6900405883789, 170.35116958618164, 117.17324161529541, 170.9763307571411, 209.35604572296143 ], "p_value": 0.0081, "mde": 30.622178816214802, "mde_rel_pct": 386.0616015800278, "verdict": "idea worse (significant)", "system_worked": false }, "mechanism_signature": { "description": "trained-model checkpoint audit signature", "signature": { "predicted": "validation loss can be flat while audited decision drift is nonzero", "observed": [ { "max_drift": 3.82235586643219, "val_range": 218.36089038848877, "audits": 20 }, { "max_drift": 3.554457426071167, "val_range": 212.48634433746338, "audits": 20 }, { "max_drift": 5.0, "val_range": 216.55184841156006, "audits": 20 }, { "max_drift": 5.0, "val_range": 200.55087280273438, "audits": 20 }, { "max_drift": 3.406560003757477, "val_range": 216.53733348846436, "audits": 20 }, { "max_drift": 3.6925817728042603, "val_range": 209.86744689941406, "audits": 20 }, { "max_drift": 3.112291097640991, "val_range": 202.15625095367432, "audits": 20 }, { "max_drift": 5.0, "val_range": 214.54338932037354, "audits": 20 } ], "mean_max_drift": 4.073530770838261, "mean_val_range": 211.3817970752716, "confirmed": false } } }