{
  "bench_version": 1,
  "track": "tabular",
  "model": "mlp",
  "metric_direction": "lower is better",
  "n_seeds": 8,
  "baseline": {
    "best_cfg": {
      "lr": 0.01
    },
    "sweep": [
      {
        "cfg": {
          "lr": 0.001
        },
        "mean": 17.006587505340576
      },
      {
        "cfg": {
          "lr": 0.003
        },
        "mean": 11.705099105834961
      },
      {
        "cfg": {
          "lr": 0.01
        },
        "mean": 7.426158666610718
      }
    ],
    "full": {
      "mean": 7.655009210109711,
      "std": 0.7549298959102074,
      "per_seed": [
        7.125377178192139,
        6.509033203125,
        7.528863430023193,
        8.541360855102539,
        7.81464958190918,
        6.914392948150635,
        8.888213157653809,
        7.918183326721191
      ],
      "n": 8
    }
  },
  "idea": {
    "mean": 7.631358027458191,
    "std": 0.8269721813703816,
    "per_seed": [
      7.135181903839111,
      6.160409450531006,
      7.647706031799316,
      8.540335655212402,
      7.844282150268555,
      6.913455486297607,
      8.89346694946289,
      7.916026592254639
    ],
    "n": 8
  },
  "comparison": {
    "delta_mean": -0.023651182651519775,
    "idea_wins": 4,
    "n_pairs": 8,
    "per_seed_diffs": [
      0.009804725646972656,
      -0.34862375259399414,
      0.11884260177612305,
      -0.0010251998901367188,
      0.029632568359375,
      -0.0009374618530273438,
      0.005253791809082031,
      -0.0021567344665527344
    ],
    "p_value": 0.9445,
    "mde": 0.1149067142918632,
    "mde_rel_pct": 1.5010656569832703,
    "verdict": "no measurable effect",
    "system_worked": false
  },
  "track_reason": "tabular is the registered optimizer/regularizer track; both systems share the same MLP and learned bottleneck",
  "idea_sweep": [
    {
      "cfg": {
        "lr": 0.001,
        "lambda": 0.03
      },
      "mean_first4": 17.134891271591187
    },
    {
      "cfg": {
        "lr": 0.001,
        "lambda": 0.1
      },
      "mean_first4": 17.241438150405884
    },
    {
      "cfg": {
        "lr": 0.001,
        "lambda": 0.3
      },
      "mean_first4": 17.401116609573364
    },
    {
      "cfg": {
        "lr": 0.003,
        "lambda": 0.03
      },
      "mean_first4": 11.699760913848877
    },
    {
      "cfg": {
        "lr": 0.003,
        "lambda": 0.1
      },
      "mean_first4": 11.717785120010376
    },
    {
      "cfg": {
        "lr": 0.003,
        "lambda": 0.3
      },
      "mean_first4": 11.66649079322815
    },
    {
      "cfg": {
        "lr": 0.01,
        "lambda": 0.03
      },
      "mean_first4": 7.370908260345459
    },
    {
      "cfg": {
        "lr": 0.01,
        "lambda": 0.1
      },
      "mean_first4": 7.376788854598999
    },
    {
      "cfg": {
        "lr": 0.01,
        "lambda": 0.3
      },
      "mean_first4": 7.378063797950745
    }
  ],
  "selected_idea_cfg": {
    "lr": 0.01,
    "lambda": 0.03
  },
  "mechanism_signature": {
    "predicted_centered_distance_correlation": "no universal value for learned encoder",
    "observed_centered_distance_correlation": 0.9999756813049316,
    "observed_centered_variance_ratio": 98.05452728271484,
    "predicted_variance_ceiling": 1.0,
    "confirmed": false
  },
  "budget": {
    "epochs": 24,
    "n_train": 400,
    "n_test": 200,
    "batch": 128,
    "hidden_dim": 32,
    "bottleneck_dim": 8,
    "baseline_grid": [
      0.001,
      0.003,
      0.01
    ],
    "idea_lambdas": [
      0.03,
      0.1,
      0.3
    ],
    "seeds": [
      0,
      1,
      2,
      3,
      4,
      5,
      6,
      7
    ]
  }
}