Bifurcation-calibrated delayed-gradient escape / bench_report.json

Failed on benchmark

Raw ⬇ ZIP
  1{
  2  "bench_version": 1,
  3  "track": "dynamics",
  4  "model": "rnn_small",
  5  "metric_direction": "lower is better",
  6  "n_seeds": 8,
  7  "baseline": {
  8    "best_cfg": {
  9      "lr": 0.01
 10    },
 11    "sweep": [
 12      {
 13        "cfg": {
 14          "lr": 0.001
 15        },
 16        "mean": 0.001222553226398304
 17      },
 18      {
 19        "cfg": {
 20          "lr": 0.003
 21        },
 22        "mean": 0.000683717749780044
 23      },
 24      {
 25        "cfg": {
 26          "lr": 0.01
 27        },
 28        "mean": 0.00021919916616752744
 29      }
 30    ],
 31    "full": {
 32      "mean": 0.0002364131532885949,
 33      "std": 0.00010524220575938558,
 34      "per_seed": [
 35        0.00023485766723752022,
 36        0.0002715944719966501,
 37        0.00015523760521318763,
 38        0.0002151069202227518,
 39        0.00015697270282544196,
 40        0.00027127040084451437,
 41        0.0001109737713704817,
 42        0.0004752916865982115
 43      ],
 44      "n": 8
 45    },
 46    "idea_grid": [
 47      {
 48        "cfg": {
 49          "lr": 0.001
 50        },
 51        "mean": 0.035339713853318244
 52      },
 53      {
 54        "cfg": {
 55          "lr": 0.003
 56        },
 57        "mean": 0.008510864296113141
 58      },
 59      {
 60        "cfg": {
 61          "lr": 0.01
 62        },
 63        "mean": 0.0002731473405219731
 64      }
 65    ]
 66  },
 67  "idea": {
 68    "mean": 0.0002731473405219731,
 69    "std": 0.00011425463363936107,
 70    "per_seed": [
 71      0.00023485766723752022,
 72      0.0002715944719966501,
 73      0.00015523760521318763,
 74      0.00041881523793563247,
 75      0.0002471378829795867,
 76      0.00027127040084451437,
 77      0.0001109737713704817,
 78      0.0004752916865982115
 79    ],
 80    "n": 8
 81  },
 82  "comparison": {
 83    "delta_mean": 3.673418723337818e-05,
 84    "idea_wins": 0,
 85    "n_pairs": 8,
 86    "per_seed_diffs": [
 87      0.0,
 88      0.0,
 89      0.0,
 90      0.00020370831771288067,
 91      9.016518015414476e-05,
 92      0.0,
 93      0.0,
 94      0.0
 95    ],
 96    "p_value": 0.5022,
 97    "mde": 6.227732731962294e-05,
 98    "mde_rel_pct": 26.342581389115686,
 99    "verdict": "no significant win",
100    "system_worked": false
101  },
102  "mechanism_signature": {
103    "mechanism_signature": {
104      "prediction": "plateau-triggered delay amplifies parameter displacement",
105      "predicted_delay_steps": 14.049755225604105,
106      "observed_delay_steps": 12,
107      "baseline_max_displacement": 6.033182621002197,
108      "idea_max_displacement": 6.033182621002197,
109      "displacement_ratio": 0.9999999999998342,
110      "k_proxy": 0.1229826378985967,
111      "confirmed": false
112    }
113  },
114  "custom_track": null
115}