Bellman-Resolvent Uncertainty Targets / results.json

Failed on benchmark

Raw ⬇ ZIP
  1{
  2  "seed": 123,
  3  "toy": {
  4    "gain_sweep": [
  5      {
  6        "lambda": 0.0,
  7        "observed_gain": 1.0,
  8        "predicted_gain": 1.0,
  9        "relative_error": 0.0
 10      },
 11      {
 12        "lambda": 0.2,
 13        "observed_gain": 1.2195121951219512,
 14        "predicted_gain": 1.2195121951219512,
 15        "relative_error": 0.0
 16      },
 17      {
 18        "lambda": 0.5,
 19        "observed_gain": 1.8181818181818181,
 20        "predicted_gain": 1.8181818181818181,
 21        "relative_error": 0.0
 22      },
 23      {
 24        "lambda": 0.8,
 25        "observed_gain": 3.571428571428572,
 26        "predicted_gain": 3.5714285714285725,
 27        "relative_error": 1.243449787580175e-16
 28      },
 29      {
 30        "lambda": 0.95,
 31        "observed_gain": 6.896551724137926,
 32        "predicted_gain": 6.896551724137931,
 33        "relative_error": 6.439293542825909e-16
 34      },
 35      {
 36        "lambda": 1.0,
 37        "observed_gain": 9.999999999999995,
 38        "predicted_gain": 10.000000000000002,
 39        "relative_error": 7.105427357601e-16
 40      }
 41    ],
 42    "truncation_sweep": [
 43      {
 44        "K": 1,
 45        "observed_residual": 2.5714285714285725,
 46        "predicted_residual": 2.5714285714285725,
 47        "relative_error": 0.0
 48      },
 49      {
 50        "K": 2,
 51        "observed_residual": 1.8514285714285723,
 52        "predicted_residual": 1.8514285714285723,
 53        "relative_error": 0.0
 54      },
 55      {
 56        "K": 4,
 57        "observed_residual": 0.9597805714285719,
 58        "predicted_residual": 0.9597805714285722,
 59        "relative_error": 3.4702401497021146e-16
 60      },
 61      {
 62        "K": 8,
 63        "observed_residual": 0.257930048681692,
 64        "predicted_residual": 0.2579300486816918,
 65        "relative_error": 8.608714109113117e-16
 66      },
 67      {
 68        "K": 12,
 69        "observed_residual": 0.0693157498634478,
 70        "predicted_residual": 0.06931574986344777,
 71        "relative_error": 4.004223523558135e-16
 72      },
 73      {
 74        "K": 20,
 75        "observed_residual": 0.005006012126272719,
 76        "predicted_residual": 0.0050060121262723726,
 77        "relative_error": 6.913234021969352e-14
 78      }
 79    ],
 80    "boundary_sweep": [
 81      {
 82        "lambda": 0.8,
 83        "gamma_lambda": 0.7200000000000001,
 84        "abs_x_K5_K10_K20": [
 85          2.8803865600000007,
 86          3.4377176341919835,
 87          3.5664225593023
 88        ],
 89        "predicted": "bounded"
 90      },
 91      {
 92        "lambda": 1.0,
 93        "gamma_lambda": 0.9,
 94        "abs_x_K5_K10_K20": [
 95          4.0951,
 96          6.5132155990000005,
 97          8.784233454094307
 98        ],
 99        "predicted": "bounded"
100      },
101      {
102        "lambda": 1.05,
103        "gamma_lambda": 0.9450000000000001,
104        "abs_x_K5_K10_K20": [
105          4.479427275625,
106          7.8552647717812745,
107          12.316744388649218
108        ],
109        "predicted": "bounded"
110      },
111      {
112        "lambda": 1.2,
113        "gamma_lambda": 1.08,
114        "abs_x_K5_K10_K20": [
115          5.86660096,
116          14.486562465909833,
117          45.76196429811626
118        ],
119        "predicted": "diverges"
120      }
121    ],
122    "variance_check": {
123      "lambda": 0.7,
124      "observed_variance": 0.8946901752274206,
125      "predicted_variance": 0.8948137326515704,
126      "relative_error": 0.00013808172543763864
127    }
128  },
129  "offline": {
130    "baseline": {
131      "value_rmse": 1.8460881177697024,
132      "bellman_rmse": 0.12847792613078712,
133      "late_state_rmse": 2.2479310635989167
134    },
135    "resolvent_weighted": {
136      "value_rmse": 1.6025952283335692,
137      "bellman_rmse": 0.12443427375170382,
138      "late_state_rmse": 1.9510774417429593
139    },
140    "uncertainty_error_correlation": 0.6106702319459555,
141    "mean_uncertainty_early": 2.512851965994944,
142    "mean_uncertainty_late": 6.532947358020794,
143    "dataset_count_early": 1728,
144    "dataset_count_late": 432
145  }
146}