anhtld commited on
Commit
27a059b
·
verified ·
1 Parent(s): ce33506

Add nonlinear chart-compat K16 selector diagnostics

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/command.txt +1 -0
  2. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/config.yaml +15 -0
  3. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/data_hash.txt +7 -0
  4. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/eval.log +3 -0
  5. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/git_hash.txt +1 -0
  6. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/metrics.json +0 -0
  7. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/metrics_by_seed.json +2762 -0
  8. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/metrics_by_task.json +117 -0
  9. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/report.md +21 -0
  10. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/split_hash.txt +20 -0
  11. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/table.tex +8 -0
  12. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/train.log +5 -0
  13. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/command.txt +1 -0
  14. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/config.yaml +15 -0
  15. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/data_hash.txt +7 -0
  16. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/eval.log +3 -0
  17. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/git_hash.txt +1 -0
  18. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/metrics.json +0 -0
  19. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/metrics_by_seed.json +2762 -0
  20. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/metrics_by_task.json +117 -0
  21. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/report.md +21 -0
  22. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/split_hash.txt +20 -0
  23. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/table.tex +8 -0
  24. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/train.log +5 -0
  25. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/command.txt +1 -0
  26. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/config.yaml +15 -0
  27. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/data_hash.txt +7 -0
  28. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/eval.log +3 -0
  29. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/git_hash.txt +1 -0
  30. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/metrics.json +0 -0
  31. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/metrics_by_seed.json +2762 -0
  32. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/metrics_by_task.json +117 -0
  33. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/report.md +21 -0
  34. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/split_hash.txt +20 -0
  35. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/table.tex +8 -0
  36. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/train.log +5 -0
  37. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/command.txt +1 -0
  38. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/config.yaml +15 -0
  39. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/data_hash.txt +7 -0
  40. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/eval.log +3 -0
  41. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/git_hash.txt +1 -0
  42. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/metrics.json +0 -0
  43. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/metrics_by_seed.json +2762 -0
  44. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/metrics_by_task.json +117 -0
  45. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/report.md +21 -0
  46. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/split_hash.txt +20 -0
  47. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/table.tex +8 -0
  48. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/train.log +5 -0
  49. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_utility_envclip_k16_train_to_test/command.txt +1 -0
  50. workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_utility_envclip_k16_train_to_test/config.yaml +15 -0
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/command.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ python scripts/eval_nonlinear_dominance_selector.py --calibration-input runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json --calibration-target-index data/cil_charts_rgb_refs/train/index.json --eval-input runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json --eval-target-index data/cil_charts_rgb_refs/test/index.json --source-index data/cil_charts_rgb_refs/train/index.json --checkpoint-template runs/ctt_residual_base_context_obs_seed{seed}/model.pt --out-dir runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test --feature-set chart_source_compat --selector-chart-feature-mode base_context_obs --target utility_margin --k 16 --bootstrap-samples 1000
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/config.yaml ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ bootstrap_samples: 1000
2
+ calibration_input: runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json
3
+ calibration_target_index: data/cil_charts_rgb_refs/train/index.json
4
+ checkpoint_template: runs/ctt_residual_base_context_obs_seed{seed}/model.pt
5
+ eval_input: runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json
6
+ eval_target_index: data/cil_charts_rgb_refs/test/index.json
7
+ feature_set: chart_source_compat
8
+ k: 16
9
+ model_types: hgb_classifier,hgb_regressor,rf_regressor
10
+ out_dir: runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test
11
+ seed: 0
12
+ selection_frac: 0.35
13
+ selector_chart_feature_mode: base_context_obs
14
+ source_index: data/cil_charts_rgb_refs/train/index.json
15
+ target: utility_margin
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/data_hash.txt ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "calibration_input": "d79e275b896eb98caa128470d4614b76960cab12407faa157c583224be7abe80",
3
+ "calibration_target_index": "67973c50072aa87a3b97533b6292b08a8b4d88a95571dc22e91b6e8fd19e0d85",
4
+ "eval_input": "541d5a928da5f7ee04ca3dea570e04208830dffde431edfe5b5e38b692634045",
5
+ "eval_target_index": "dee1b5225485df6624747dd4c5074bcf514685b54017d05e698d2051638fbf36",
6
+ "source_index": "67973c50072aa87a3b97533b6292b08a8b4d88a95571dc22e91b6e8fd19e0d85"
7
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/eval.log ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ selected model/tau on held-out calibration-selection rows; evaluated on held-out measured rollout rows
2
+ eval_input=runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json
3
+ num_eval_rows=144
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/git_hash.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ 14e81292e2004925f81e0bb4810f6f144bfd87b7
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/metrics.json ADDED
The diff for this file is too large to render. See raw diff
 
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/metrics_by_seed.json ADDED
@@ -0,0 +1,2762 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "1073161778/train0": {
3
+ "base_success": 1.0,
4
+ "base_utility": 2.0,
5
+ "coverage": 1.0,
6
+ "execute_generated": 1.0,
7
+ "fallback_rate": 0.0,
8
+ "hidden_chart_oracle_success": 1.0,
9
+ "hidden_chart_oracle_utility": 2.0,
10
+ "outcome_ptr": 0.0,
11
+ "predicted_margin": 0.34177985507477204,
12
+ "proposal_oracle_success": 1.0,
13
+ "proposal_oracle_utility": 2.0,
14
+ "selected_candidate_index": 6.0,
15
+ "selected_success": 0.0,
16
+ "selected_success_gain_over_base": -1.0,
17
+ "selected_utility": 0.67014479637146,
18
+ "selected_utility_gain_over_base": -1.32985520362854,
19
+ "selector_regret": 1.32985520362854,
20
+ "success_selector_gap": 1.0,
21
+ "success_support_gap": 0.0,
22
+ "support_gap": 0.0,
23
+ "tau": 0.29031104002937436
24
+ },
25
+ "1073161778/train1": {
26
+ "base_success": 1.0,
27
+ "base_utility": 2.0,
28
+ "coverage": 1.0,
29
+ "execute_generated": 1.0,
30
+ "fallback_rate": 0.0,
31
+ "hidden_chart_oracle_success": 1.0,
32
+ "hidden_chart_oracle_utility": 2.0,
33
+ "outcome_ptr": 0.0,
34
+ "predicted_margin": 0.4119603552284239,
35
+ "proposal_oracle_success": 1.0,
36
+ "proposal_oracle_utility": 2.0,
37
+ "selected_candidate_index": 7.0,
38
+ "selected_success": 0.0,
39
+ "selected_success_gain_over_base": -1.0,
40
+ "selected_utility": 0.6916947960853577,
41
+ "selected_utility_gain_over_base": -1.3083052039146423,
42
+ "selector_regret": 1.3083052039146423,
43
+ "success_selector_gap": 1.0,
44
+ "success_support_gap": 0.0,
45
+ "support_gap": 0.0,
46
+ "tau": 0.29031104002937436
47
+ },
48
+ "1073161778/train2": {
49
+ "base_success": 1.0,
50
+ "base_utility": 2.0,
51
+ "coverage": 1.0,
52
+ "execute_generated": 1.0,
53
+ "fallback_rate": 0.0,
54
+ "hidden_chart_oracle_success": 1.0,
55
+ "hidden_chart_oracle_utility": 2.0,
56
+ "outcome_ptr": 0.0,
57
+ "predicted_margin": 0.33062341715384946,
58
+ "proposal_oracle_success": 1.0,
59
+ "proposal_oracle_utility": 2.0,
60
+ "selected_candidate_index": 6.0,
61
+ "selected_success": 1.0,
62
+ "selected_success_gain_over_base": 0.0,
63
+ "selected_utility": 2.0,
64
+ "selected_utility_gain_over_base": 0.0,
65
+ "selector_regret": 0.0,
66
+ "success_selector_gap": 0.0,
67
+ "success_support_gap": 0.0,
68
+ "support_gap": 0.0,
69
+ "tau": 0.29031104002937436
70
+ },
71
+ "1113371261/train0": {
72
+ "base_success": 0.0,
73
+ "base_utility": 0.15842342376708984,
74
+ "coverage": 0.0,
75
+ "execute_generated": 0.0,
76
+ "fallback_rate": 1.0,
77
+ "hidden_chart_oracle_success": 1.0,
78
+ "hidden_chart_oracle_utility": 2.0,
79
+ "outcome_ptr": 0.0,
80
+ "predicted_margin": 0.2414028495277438,
81
+ "proposal_oracle_success": 0.0,
82
+ "proposal_oracle_utility": 0.10274861752986908,
83
+ "selected_candidate_index": 0.0,
84
+ "selected_success": 0.0,
85
+ "selected_success_gain_over_base": 0.0,
86
+ "selected_utility": 0.15842342376708984,
87
+ "selected_utility_gain_over_base": 0.0,
88
+ "selector_regret": 0.0,
89
+ "success_selector_gap": 0.0,
90
+ "success_support_gap": 1.0,
91
+ "support_gap": 1.897251382470131,
92
+ "tau": 0.29031104002937436
93
+ },
94
+ "1113371261/train1": {
95
+ "base_success": 0.0,
96
+ "base_utility": 0.15842342376708984,
97
+ "coverage": 0.0,
98
+ "execute_generated": 0.0,
99
+ "fallback_rate": 1.0,
100
+ "hidden_chart_oracle_success": 1.0,
101
+ "hidden_chart_oracle_utility": 2.0,
102
+ "outcome_ptr": 1.0,
103
+ "predicted_margin": 0.25784951374924575,
104
+ "proposal_oracle_success": 0.0,
105
+ "proposal_oracle_utility": 0.18093222379684448,
106
+ "selected_candidate_index": 4.0,
107
+ "selected_success": 0.0,
108
+ "selected_success_gain_over_base": 0.0,
109
+ "selected_utility": 0.15842342376708984,
110
+ "selected_utility_gain_over_base": 0.0,
111
+ "selector_regret": 0.02250880002975464,
112
+ "success_selector_gap": 0.0,
113
+ "success_support_gap": 1.0,
114
+ "support_gap": 1.8190677762031555,
115
+ "tau": 0.29031104002937436
116
+ },
117
+ "1113371261/train2": {
118
+ "base_success": 0.0,
119
+ "base_utility": 0.15842342376708984,
120
+ "coverage": 0.0,
121
+ "execute_generated": 0.0,
122
+ "fallback_rate": 1.0,
123
+ "hidden_chart_oracle_success": 1.0,
124
+ "hidden_chart_oracle_utility": 2.0,
125
+ "outcome_ptr": 1.0,
126
+ "predicted_margin": 0.2414028495277438,
127
+ "proposal_oracle_success": 0.0,
128
+ "proposal_oracle_utility": 0.19527167081832886,
129
+ "selected_candidate_index": 2.0,
130
+ "selected_success": 0.0,
131
+ "selected_success_gain_over_base": 0.0,
132
+ "selected_utility": 0.15842342376708984,
133
+ "selected_utility_gain_over_base": 0.0,
134
+ "selector_regret": 0.036848247051239014,
135
+ "success_selector_gap": 0.0,
136
+ "success_support_gap": 1.0,
137
+ "support_gap": 1.8047283291816711,
138
+ "tau": 0.29031104002937436
139
+ },
140
+ "1151410346/train0": {
141
+ "base_success": 0.0,
142
+ "base_utility": 0.2260012924671173,
143
+ "coverage": 0.0,
144
+ "execute_generated": 0.0,
145
+ "fallback_rate": 1.0,
146
+ "hidden_chart_oracle_success": 1.0,
147
+ "hidden_chart_oracle_utility": 2.0,
148
+ "outcome_ptr": 1.0,
149
+ "predicted_margin": 0.19860467482965133,
150
+ "proposal_oracle_success": 0.0,
151
+ "proposal_oracle_utility": 0.8663521409034729,
152
+ "selected_candidate_index": 13.0,
153
+ "selected_success": 0.0,
154
+ "selected_success_gain_over_base": 0.0,
155
+ "selected_utility": 0.2260012924671173,
156
+ "selected_utility_gain_over_base": 0.0,
157
+ "selector_regret": 0.6403508484363556,
158
+ "success_selector_gap": 0.0,
159
+ "success_support_gap": 1.0,
160
+ "support_gap": 1.133647859096527,
161
+ "tau": 0.29031104002937436
162
+ },
163
+ "1151410346/train1": {
164
+ "base_success": 0.0,
165
+ "base_utility": 0.2260012924671173,
166
+ "coverage": 0.0,
167
+ "execute_generated": 0.0,
168
+ "fallback_rate": 1.0,
169
+ "hidden_chart_oracle_success": 1.0,
170
+ "hidden_chart_oracle_utility": 2.0,
171
+ "outcome_ptr": 1.0,
172
+ "predicted_margin": 0.20195660070480034,
173
+ "proposal_oracle_success": 1.0,
174
+ "proposal_oracle_utility": 2.0,
175
+ "selected_candidate_index": 9.0,
176
+ "selected_success": 0.0,
177
+ "selected_success_gain_over_base": 0.0,
178
+ "selected_utility": 0.2260012924671173,
179
+ "selected_utility_gain_over_base": 0.0,
180
+ "selector_regret": 1.7739987075328827,
181
+ "success_selector_gap": 1.0,
182
+ "success_support_gap": 0.0,
183
+ "support_gap": 0.0,
184
+ "tau": 0.29031104002937436
185
+ },
186
+ "1151410346/train2": {
187
+ "base_success": 0.0,
188
+ "base_utility": 0.2260012924671173,
189
+ "coverage": 0.0,
190
+ "execute_generated": 0.0,
191
+ "fallback_rate": 1.0,
192
+ "hidden_chart_oracle_success": 1.0,
193
+ "hidden_chart_oracle_utility": 2.0,
194
+ "outcome_ptr": 1.0,
195
+ "predicted_margin": 0.20911211212626157,
196
+ "proposal_oracle_success": 0.0,
197
+ "proposal_oracle_utility": 0.5522047281265259,
198
+ "selected_candidate_index": 11.0,
199
+ "selected_success": 0.0,
200
+ "selected_success_gain_over_base": 0.0,
201
+ "selected_utility": 0.2260012924671173,
202
+ "selected_utility_gain_over_base": 0.0,
203
+ "selector_regret": 0.32620343565940857,
204
+ "success_selector_gap": 0.0,
205
+ "success_support_gap": 1.0,
206
+ "support_gap": 1.4477952718734741,
207
+ "tau": 0.29031104002937436
208
+ },
209
+ "1261930456/train0": {
210
+ "base_success": 1.0,
211
+ "base_utility": 2.0,
212
+ "coverage": 1.0,
213
+ "execute_generated": 1.0,
214
+ "fallback_rate": 0.0,
215
+ "hidden_chart_oracle_success": 1.0,
216
+ "hidden_chart_oracle_utility": 2.0,
217
+ "outcome_ptr": 0.0,
218
+ "predicted_margin": 0.38402002330650076,
219
+ "proposal_oracle_success": 1.0,
220
+ "proposal_oracle_utility": 2.0,
221
+ "selected_candidate_index": 5.0,
222
+ "selected_success": 1.0,
223
+ "selected_success_gain_over_base": 0.0,
224
+ "selected_utility": 2.0,
225
+ "selected_utility_gain_over_base": 0.0,
226
+ "selector_regret": 0.0,
227
+ "success_selector_gap": 0.0,
228
+ "success_support_gap": 0.0,
229
+ "support_gap": 0.0,
230
+ "tau": 0.29031104002937436
231
+ },
232
+ "1261930456/train1": {
233
+ "base_success": 1.0,
234
+ "base_utility": 2.0,
235
+ "coverage": 1.0,
236
+ "execute_generated": 1.0,
237
+ "fallback_rate": 0.0,
238
+ "hidden_chart_oracle_success": 1.0,
239
+ "hidden_chart_oracle_utility": 2.0,
240
+ "outcome_ptr": 0.0,
241
+ "predicted_margin": 0.4863910843674545,
242
+ "proposal_oracle_success": 1.0,
243
+ "proposal_oracle_utility": 2.0,
244
+ "selected_candidate_index": 11.0,
245
+ "selected_success": 0.0,
246
+ "selected_success_gain_over_base": -1.0,
247
+ "selected_utility": 0.08180709183216095,
248
+ "selected_utility_gain_over_base": -1.918192908167839,
249
+ "selector_regret": 1.918192908167839,
250
+ "success_selector_gap": 1.0,
251
+ "success_support_gap": 0.0,
252
+ "support_gap": 0.0,
253
+ "tau": 0.29031104002937436
254
+ },
255
+ "1261930456/train2": {
256
+ "base_success": 1.0,
257
+ "base_utility": 2.0,
258
+ "coverage": 1.0,
259
+ "execute_generated": 1.0,
260
+ "fallback_rate": 0.0,
261
+ "hidden_chart_oracle_success": 1.0,
262
+ "hidden_chart_oracle_utility": 2.0,
263
+ "outcome_ptr": 0.0,
264
+ "predicted_margin": 0.47661080026664676,
265
+ "proposal_oracle_success": 1.0,
266
+ "proposal_oracle_utility": 2.0,
267
+ "selected_candidate_index": 2.0,
268
+ "selected_success": 1.0,
269
+ "selected_success_gain_over_base": 0.0,
270
+ "selected_utility": 2.0,
271
+ "selected_utility_gain_over_base": 0.0,
272
+ "selector_regret": 0.0,
273
+ "success_selector_gap": 0.0,
274
+ "success_support_gap": 0.0,
275
+ "support_gap": 0.0,
276
+ "tau": 0.29031104002937436
277
+ },
278
+ "1324133122/train0": {
279
+ "base_success": 0.0,
280
+ "base_utility": 0.09220491349697113,
281
+ "coverage": 0.0,
282
+ "execute_generated": 0.0,
283
+ "fallback_rate": 1.0,
284
+ "hidden_chart_oracle_success": 1.0,
285
+ "hidden_chart_oracle_utility": 2.0,
286
+ "outcome_ptr": 0.0,
287
+ "predicted_margin": 0.15883484058846561,
288
+ "proposal_oracle_success": 0.0,
289
+ "proposal_oracle_utility": 0.054989203810691833,
290
+ "selected_candidate_index": 0.0,
291
+ "selected_success": 0.0,
292
+ "selected_success_gain_over_base": 0.0,
293
+ "selected_utility": 0.09220491349697113,
294
+ "selected_utility_gain_over_base": 0.0,
295
+ "selector_regret": 0.0,
296
+ "success_selector_gap": 0.0,
297
+ "success_support_gap": 1.0,
298
+ "support_gap": 1.9450107961893082,
299
+ "tau": 0.29031104002937436
300
+ },
301
+ "1324133122/train1": {
302
+ "base_success": 0.0,
303
+ "base_utility": 0.09220491349697113,
304
+ "coverage": 0.0,
305
+ "execute_generated": 0.0,
306
+ "fallback_rate": 1.0,
307
+ "hidden_chart_oracle_success": 1.0,
308
+ "hidden_chart_oracle_utility": 2.0,
309
+ "outcome_ptr": 1.0,
310
+ "predicted_margin": 0.10585396767858887,
311
+ "proposal_oracle_success": 0.0,
312
+ "proposal_oracle_utility": 0.1324586570262909,
313
+ "selected_candidate_index": 1.0,
314
+ "selected_success": 0.0,
315
+ "selected_success_gain_over_base": 0.0,
316
+ "selected_utility": 0.09220491349697113,
317
+ "selected_utility_gain_over_base": 0.0,
318
+ "selector_regret": 0.04025374352931976,
319
+ "success_selector_gap": 0.0,
320
+ "success_support_gap": 1.0,
321
+ "support_gap": 1.867541342973709,
322
+ "tau": 0.29031104002937436
323
+ },
324
+ "1324133122/train2": {
325
+ "base_success": 0.0,
326
+ "base_utility": 0.09220491349697113,
327
+ "coverage": 0.0,
328
+ "execute_generated": 0.0,
329
+ "fallback_rate": 1.0,
330
+ "hidden_chart_oracle_success": 1.0,
331
+ "hidden_chart_oracle_utility": 2.0,
332
+ "outcome_ptr": 1.0,
333
+ "predicted_margin": 0.16360521054531033,
334
+ "proposal_oracle_success": 0.0,
335
+ "proposal_oracle_utility": 0.19291165471076965,
336
+ "selected_candidate_index": 0.0,
337
+ "selected_success": 0.0,
338
+ "selected_success_gain_over_base": 0.0,
339
+ "selected_utility": 0.09220491349697113,
340
+ "selected_utility_gain_over_base": 0.0,
341
+ "selector_regret": 0.10070674121379852,
342
+ "success_selector_gap": 0.0,
343
+ "success_support_gap": 1.0,
344
+ "support_gap": 1.8070883452892303,
345
+ "tau": 0.29031104002937436
346
+ },
347
+ "1324642411/train0": {
348
+ "base_success": 0.0,
349
+ "base_utility": 0.10647052526473999,
350
+ "coverage": 1.0,
351
+ "execute_generated": 1.0,
352
+ "fallback_rate": 0.0,
353
+ "hidden_chart_oracle_success": 1.0,
354
+ "hidden_chart_oracle_utility": 2.0,
355
+ "outcome_ptr": 1.0,
356
+ "predicted_margin": 0.4193019968970795,
357
+ "proposal_oracle_success": 0.0,
358
+ "proposal_oracle_utility": 0.14406004548072815,
359
+ "selected_candidate_index": 12.0,
360
+ "selected_success": 0.0,
361
+ "selected_success_gain_over_base": 0.0,
362
+ "selected_utility": 0.14406004548072815,
363
+ "selected_utility_gain_over_base": 0.03758952021598816,
364
+ "selector_regret": 0.0,
365
+ "success_selector_gap": 0.0,
366
+ "success_support_gap": 1.0,
367
+ "support_gap": 1.8559399545192719,
368
+ "tau": 0.29031104002937436
369
+ },
370
+ "1324642411/train1": {
371
+ "base_success": 0.0,
372
+ "base_utility": 0.10647052526473999,
373
+ "coverage": 1.0,
374
+ "execute_generated": 1.0,
375
+ "fallback_rate": 0.0,
376
+ "hidden_chart_oracle_success": 1.0,
377
+ "hidden_chart_oracle_utility": 2.0,
378
+ "outcome_ptr": 1.0,
379
+ "predicted_margin": 0.36738128595033803,
380
+ "proposal_oracle_success": 1.0,
381
+ "proposal_oracle_utility": 2.0,
382
+ "selected_candidate_index": 9.0,
383
+ "selected_success": 0.0,
384
+ "selected_success_gain_over_base": 0.0,
385
+ "selected_utility": 0.1280023753643036,
386
+ "selected_utility_gain_over_base": 0.0215318500995636,
387
+ "selector_regret": 1.8719976246356964,
388
+ "success_selector_gap": 1.0,
389
+ "success_support_gap": 0.0,
390
+ "support_gap": 0.0,
391
+ "tau": 0.29031104002937436
392
+ },
393
+ "1324642411/train2": {
394
+ "base_success": 0.0,
395
+ "base_utility": 0.10647052526473999,
396
+ "coverage": 1.0,
397
+ "execute_generated": 1.0,
398
+ "fallback_rate": 0.0,
399
+ "hidden_chart_oracle_success": 1.0,
400
+ "hidden_chart_oracle_utility": 2.0,
401
+ "outcome_ptr": 1.0,
402
+ "predicted_margin": 0.445409885616724,
403
+ "proposal_oracle_success": 1.0,
404
+ "proposal_oracle_utility": 2.0,
405
+ "selected_candidate_index": 14.0,
406
+ "selected_success": 0.0,
407
+ "selected_success_gain_over_base": 0.0,
408
+ "selected_utility": 0.1215444803237915,
409
+ "selected_utility_gain_over_base": 0.015073955059051514,
410
+ "selector_regret": 1.8784555196762085,
411
+ "success_selector_gap": 1.0,
412
+ "success_support_gap": 0.0,
413
+ "support_gap": 0.0,
414
+ "tau": 0.29031104002937436
415
+ },
416
+ "1375492786/train0": {
417
+ "base_success": 1.0,
418
+ "base_utility": 2.0,
419
+ "coverage": 0.0,
420
+ "execute_generated": 0.0,
421
+ "fallback_rate": 1.0,
422
+ "hidden_chart_oracle_success": 1.0,
423
+ "hidden_chart_oracle_utility": 2.0,
424
+ "outcome_ptr": 0.0,
425
+ "predicted_margin": 0.11744128190863853,
426
+ "proposal_oracle_success": 1.0,
427
+ "proposal_oracle_utility": 2.0,
428
+ "selected_candidate_index": 12.0,
429
+ "selected_success": 1.0,
430
+ "selected_success_gain_over_base": 0.0,
431
+ "selected_utility": 2.0,
432
+ "selected_utility_gain_over_base": 0.0,
433
+ "selector_regret": 0.0,
434
+ "success_selector_gap": 0.0,
435
+ "success_support_gap": 0.0,
436
+ "support_gap": 0.0,
437
+ "tau": 0.29031104002937436
438
+ },
439
+ "1375492786/train1": {
440
+ "base_success": 1.0,
441
+ "base_utility": 2.0,
442
+ "coverage": 0.0,
443
+ "execute_generated": 0.0,
444
+ "fallback_rate": 1.0,
445
+ "hidden_chart_oracle_success": 1.0,
446
+ "hidden_chart_oracle_utility": 2.0,
447
+ "outcome_ptr": 0.0,
448
+ "predicted_margin": 0.11061386076837161,
449
+ "proposal_oracle_success": 1.0,
450
+ "proposal_oracle_utility": 2.0,
451
+ "selected_candidate_index": 10.0,
452
+ "selected_success": 1.0,
453
+ "selected_success_gain_over_base": 0.0,
454
+ "selected_utility": 2.0,
455
+ "selected_utility_gain_over_base": 0.0,
456
+ "selector_regret": 0.0,
457
+ "success_selector_gap": 0.0,
458
+ "success_support_gap": 0.0,
459
+ "support_gap": 0.0,
460
+ "tau": 0.29031104002937436
461
+ },
462
+ "1375492786/train2": {
463
+ "base_success": 1.0,
464
+ "base_utility": 2.0,
465
+ "coverage": 0.0,
466
+ "execute_generated": 0.0,
467
+ "fallback_rate": 1.0,
468
+ "hidden_chart_oracle_success": 1.0,
469
+ "hidden_chart_oracle_utility": 2.0,
470
+ "outcome_ptr": 0.0,
471
+ "predicted_margin": 0.1488380425829457,
472
+ "proposal_oracle_success": 1.0,
473
+ "proposal_oracle_utility": 2.0,
474
+ "selected_candidate_index": 0.0,
475
+ "selected_success": 1.0,
476
+ "selected_success_gain_over_base": 0.0,
477
+ "selected_utility": 2.0,
478
+ "selected_utility_gain_over_base": 0.0,
479
+ "selector_regret": 0.0,
480
+ "success_selector_gap": 0.0,
481
+ "success_support_gap": 0.0,
482
+ "support_gap": 0.0,
483
+ "tau": 0.29031104002937436
484
+ },
485
+ "141984615/train0": {
486
+ "base_success": 0.0,
487
+ "base_utility": 0.10283424705266953,
488
+ "coverage": 1.0,
489
+ "execute_generated": 1.0,
490
+ "fallback_rate": 0.0,
491
+ "hidden_chart_oracle_success": 0.0,
492
+ "hidden_chart_oracle_utility": 0.2857772707939148,
493
+ "outcome_ptr": 0.0,
494
+ "predicted_margin": 0.3220294334917424,
495
+ "proposal_oracle_success": 0.0,
496
+ "proposal_oracle_utility": 0.015177905559539795,
497
+ "selected_candidate_index": 4.0,
498
+ "selected_success": 0.0,
499
+ "selected_success_gain_over_base": 0.0,
500
+ "selected_utility": 0.006893515586853027,
501
+ "selected_utility_gain_over_base": -0.0959407314658165,
502
+ "selector_regret": 0.008284389972686768,
503
+ "success_selector_gap": 0.0,
504
+ "success_support_gap": 0.0,
505
+ "support_gap": 0.270599365234375,
506
+ "tau": 0.29031104002937436
507
+ },
508
+ "141984615/train1": {
509
+ "base_success": 0.0,
510
+ "base_utility": 0.10283424705266953,
511
+ "coverage": 1.0,
512
+ "execute_generated": 1.0,
513
+ "fallback_rate": 0.0,
514
+ "hidden_chart_oracle_success": 0.0,
515
+ "hidden_chart_oracle_utility": 0.2857772707939148,
516
+ "outcome_ptr": 0.0,
517
+ "predicted_margin": 0.31716416784041074,
518
+ "proposal_oracle_success": 0.0,
519
+ "proposal_oracle_utility": 0.037011127918958664,
520
+ "selected_candidate_index": 1.0,
521
+ "selected_success": 0.0,
522
+ "selected_success_gain_over_base": 0.0,
523
+ "selected_utility": 0.01966226100921631,
524
+ "selected_utility_gain_over_base": -0.08317198604345322,
525
+ "selector_regret": 0.017348866909742355,
526
+ "success_selector_gap": 0.0,
527
+ "success_support_gap": 0.0,
528
+ "support_gap": 0.24876614287495613,
529
+ "tau": 0.29031104002937436
530
+ },
531
+ "141984615/train2": {
532
+ "base_success": 0.0,
533
+ "base_utility": 0.10283424705266953,
534
+ "coverage": 1.0,
535
+ "execute_generated": 1.0,
536
+ "fallback_rate": 0.0,
537
+ "hidden_chart_oracle_success": 0.0,
538
+ "hidden_chart_oracle_utility": 0.2857772707939148,
539
+ "outcome_ptr": 0.0,
540
+ "predicted_margin": 0.3160658333588705,
541
+ "proposal_oracle_success": 0.0,
542
+ "proposal_oracle_utility": 0.058951716870069504,
543
+ "selected_candidate_index": 11.0,
544
+ "selected_success": 0.0,
545
+ "selected_success_gain_over_base": 0.0,
546
+ "selected_utility": 0.058951716870069504,
547
+ "selected_utility_gain_over_base": -0.04388253018260002,
548
+ "selector_regret": 0.0,
549
+ "success_selector_gap": 0.0,
550
+ "success_support_gap": 0.0,
551
+ "support_gap": 0.2268255539238453,
552
+ "tau": 0.29031104002937436
553
+ },
554
+ "1461069713/train0": {
555
+ "base_success": 0.0,
556
+ "base_utility": 0.0939030647277832,
557
+ "coverage": 1.0,
558
+ "execute_generated": 1.0,
559
+ "fallback_rate": 0.0,
560
+ "hidden_chart_oracle_success": 0.0,
561
+ "hidden_chart_oracle_utility": 0.1593424528837204,
562
+ "outcome_ptr": 0.0,
563
+ "predicted_margin": 0.3547798191536652,
564
+ "proposal_oracle_success": 0.0,
565
+ "proposal_oracle_utility": 0.06019836664199829,
566
+ "selected_candidate_index": 2.0,
567
+ "selected_success": 0.0,
568
+ "selected_success_gain_over_base": 0.0,
569
+ "selected_utility": 0.06019836664199829,
570
+ "selected_utility_gain_over_base": -0.03370469808578491,
571
+ "selector_regret": 0.0,
572
+ "success_selector_gap": 0.0,
573
+ "success_support_gap": 0.0,
574
+ "support_gap": 0.0991440862417221,
575
+ "tau": 0.29031104002937436
576
+ },
577
+ "1461069713/train1": {
578
+ "base_success": 0.0,
579
+ "base_utility": 0.0939030647277832,
580
+ "coverage": 1.0,
581
+ "execute_generated": 1.0,
582
+ "fallback_rate": 0.0,
583
+ "hidden_chart_oracle_success": 0.0,
584
+ "hidden_chart_oracle_utility": 0.1593424528837204,
585
+ "outcome_ptr": 0.0,
586
+ "predicted_margin": 0.6871219206697284,
587
+ "proposal_oracle_success": 0.0,
588
+ "proposal_oracle_utility": 0.016585012897849083,
589
+ "selected_candidate_index": 0.0,
590
+ "selected_success": 0.0,
591
+ "selected_success_gain_over_base": 0.0,
592
+ "selected_utility": 0.009687979705631733,
593
+ "selected_utility_gain_over_base": -0.08421508502215147,
594
+ "selector_regret": 0.00689703319221735,
595
+ "success_selector_gap": 0.0,
596
+ "success_support_gap": 0.0,
597
+ "support_gap": 0.14275743998587132,
598
+ "tau": 0.29031104002937436
599
+ },
600
+ "1461069713/train2": {
601
+ "base_success": 0.0,
602
+ "base_utility": 0.0939030647277832,
603
+ "coverage": 1.0,
604
+ "execute_generated": 1.0,
605
+ "fallback_rate": 0.0,
606
+ "hidden_chart_oracle_success": 0.0,
607
+ "hidden_chart_oracle_utility": 0.1593424528837204,
608
+ "outcome_ptr": 1.0,
609
+ "predicted_margin": 0.31492302371466396,
610
+ "proposal_oracle_success": 0.0,
611
+ "proposal_oracle_utility": 0.09997648000717163,
612
+ "selected_candidate_index": 5.0,
613
+ "selected_success": 0.0,
614
+ "selected_success_gain_over_base": 0.0,
615
+ "selected_utility": 0.06987287849187851,
616
+ "selected_utility_gain_over_base": -0.024030186235904694,
617
+ "selector_regret": 0.03010360151529312,
618
+ "success_selector_gap": 0.0,
619
+ "success_support_gap": 0.0,
620
+ "support_gap": 0.05936597287654877,
621
+ "tau": 0.29031104002937436
622
+ },
623
+ "1480856810/train0": {
624
+ "base_success": 0.0,
625
+ "base_utility": 0.08703407645225525,
626
+ "coverage": 1.0,
627
+ "execute_generated": 1.0,
628
+ "fallback_rate": 0.0,
629
+ "hidden_chart_oracle_success": 0.0,
630
+ "hidden_chart_oracle_utility": 0.20948036015033722,
631
+ "outcome_ptr": 1.0,
632
+ "predicted_margin": 0.3427743477646694,
633
+ "proposal_oracle_success": 1.0,
634
+ "proposal_oracle_utility": 2.0,
635
+ "selected_candidate_index": 1.0,
636
+ "selected_success": 0.0,
637
+ "selected_success_gain_over_base": 0.0,
638
+ "selected_utility": 0.06508849561214447,
639
+ "selected_utility_gain_over_base": -0.02194558084011078,
640
+ "selector_regret": 1.9349115043878555,
641
+ "success_selector_gap": 1.0,
642
+ "success_support_gap": 0.0,
643
+ "support_gap": 0.0,
644
+ "tau": 0.29031104002937436
645
+ },
646
+ "1480856810/train1": {
647
+ "base_success": 0.0,
648
+ "base_utility": 0.08703407645225525,
649
+ "coverage": 1.0,
650
+ "execute_generated": 1.0,
651
+ "fallback_rate": 0.0,
652
+ "hidden_chart_oracle_success": 0.0,
653
+ "hidden_chart_oracle_utility": 0.20948036015033722,
654
+ "outcome_ptr": 1.0,
655
+ "predicted_margin": 0.5935337022083772,
656
+ "proposal_oracle_success": 1.0,
657
+ "proposal_oracle_utility": 2.0,
658
+ "selected_candidate_index": 13.0,
659
+ "selected_success": 0.0,
660
+ "selected_success_gain_over_base": 0.0,
661
+ "selected_utility": 0.20681259036064148,
662
+ "selected_utility_gain_over_base": 0.11977851390838623,
663
+ "selector_regret": 1.7931874096393585,
664
+ "success_selector_gap": 1.0,
665
+ "success_support_gap": 0.0,
666
+ "support_gap": 0.0,
667
+ "tau": 0.29031104002937436
668
+ },
669
+ "1480856810/train2": {
670
+ "base_success": 0.0,
671
+ "base_utility": 0.08703407645225525,
672
+ "coverage": 1.0,
673
+ "execute_generated": 1.0,
674
+ "fallback_rate": 0.0,
675
+ "hidden_chart_oracle_success": 0.0,
676
+ "hidden_chart_oracle_utility": 0.20948036015033722,
677
+ "outcome_ptr": 1.0,
678
+ "predicted_margin": 0.3427743477646694,
679
+ "proposal_oracle_success": 1.0,
680
+ "proposal_oracle_utility": 2.0,
681
+ "selected_candidate_index": 5.0,
682
+ "selected_success": 0.0,
683
+ "selected_success_gain_over_base": 0.0,
684
+ "selected_utility": 0.0997026115655899,
685
+ "selected_utility_gain_over_base": 0.012668535113334656,
686
+ "selector_regret": 1.90029738843441,
687
+ "success_selector_gap": 1.0,
688
+ "success_support_gap": 0.0,
689
+ "support_gap": 0.0,
690
+ "tau": 0.29031104002937436
691
+ },
692
+ "1509980664/train0": {
693
+ "base_success": 0.0,
694
+ "base_utility": 0.8334100246429443,
695
+ "coverage": 1.0,
696
+ "execute_generated": 1.0,
697
+ "fallback_rate": 0.0,
698
+ "hidden_chart_oracle_success": 1.0,
699
+ "hidden_chart_oracle_utility": 2.0,
700
+ "outcome_ptr": 1.0,
701
+ "predicted_margin": 0.6195026651718122,
702
+ "proposal_oracle_success": 1.0,
703
+ "proposal_oracle_utility": 2.0,
704
+ "selected_candidate_index": 9.0,
705
+ "selected_success": 0.0,
706
+ "selected_success_gain_over_base": 0.0,
707
+ "selected_utility": 0.8125001788139343,
708
+ "selected_utility_gain_over_base": -0.02090984582901001,
709
+ "selector_regret": 1.1874998211860657,
710
+ "success_selector_gap": 1.0,
711
+ "success_support_gap": 0.0,
712
+ "support_gap": 0.0,
713
+ "tau": 0.29031104002937436
714
+ },
715
+ "1509980664/train1": {
716
+ "base_success": 0.0,
717
+ "base_utility": 0.8334100246429443,
718
+ "coverage": 1.0,
719
+ "execute_generated": 1.0,
720
+ "fallback_rate": 0.0,
721
+ "hidden_chart_oracle_success": 1.0,
722
+ "hidden_chart_oracle_utility": 2.0,
723
+ "outcome_ptr": 1.0,
724
+ "predicted_margin": 0.7082563723226508,
725
+ "proposal_oracle_success": 1.0,
726
+ "proposal_oracle_utility": 2.0,
727
+ "selected_candidate_index": 4.0,
728
+ "selected_success": 0.0,
729
+ "selected_success_gain_over_base": 0.0,
730
+ "selected_utility": 0.21804393827915192,
731
+ "selected_utility_gain_over_base": -0.6153660863637924,
732
+ "selector_regret": 1.781956061720848,
733
+ "success_selector_gap": 1.0,
734
+ "success_support_gap": 0.0,
735
+ "support_gap": 0.0,
736
+ "tau": 0.29031104002937436
737
+ },
738
+ "1509980664/train2": {
739
+ "base_success": 0.0,
740
+ "base_utility": 0.8334100246429443,
741
+ "coverage": 1.0,
742
+ "execute_generated": 1.0,
743
+ "fallback_rate": 0.0,
744
+ "hidden_chart_oracle_success": 1.0,
745
+ "hidden_chart_oracle_utility": 2.0,
746
+ "outcome_ptr": 1.0,
747
+ "predicted_margin": 0.6819725947753494,
748
+ "proposal_oracle_success": 1.0,
749
+ "proposal_oracle_utility": 2.0,
750
+ "selected_candidate_index": 12.0,
751
+ "selected_success": 0.0,
752
+ "selected_success_gain_over_base": 0.0,
753
+ "selected_utility": 0.8189421892166138,
754
+ "selected_utility_gain_over_base": -0.014467835426330566,
755
+ "selector_regret": 1.1810578107833862,
756
+ "success_selector_gap": 1.0,
757
+ "success_support_gap": 0.0,
758
+ "support_gap": 0.0,
759
+ "tau": 0.29031104002937436
760
+ },
761
+ "1522272686/train0": {
762
+ "base_success": 0.0,
763
+ "base_utility": 0.5839591026306152,
764
+ "coverage": 0.0,
765
+ "execute_generated": 0.0,
766
+ "fallback_rate": 1.0,
767
+ "hidden_chart_oracle_success": 1.0,
768
+ "hidden_chart_oracle_utility": 2.0,
769
+ "outcome_ptr": 0.0,
770
+ "predicted_margin": 0.0722437728966473,
771
+ "proposal_oracle_success": 0.0,
772
+ "proposal_oracle_utility": 0.18325337767601013,
773
+ "selected_candidate_index": 7.0,
774
+ "selected_success": 0.0,
775
+ "selected_success_gain_over_base": 0.0,
776
+ "selected_utility": 0.5839591026306152,
777
+ "selected_utility_gain_over_base": 0.0,
778
+ "selector_regret": 0.0,
779
+ "success_selector_gap": 0.0,
780
+ "success_support_gap": 1.0,
781
+ "support_gap": 1.8167466223239899,
782
+ "tau": 0.29031104002937436
783
+ },
784
+ "1522272686/train1": {
785
+ "base_success": 0.0,
786
+ "base_utility": 0.5839591026306152,
787
+ "coverage": 0.0,
788
+ "execute_generated": 0.0,
789
+ "fallback_rate": 1.0,
790
+ "hidden_chart_oracle_success": 1.0,
791
+ "hidden_chart_oracle_utility": 2.0,
792
+ "outcome_ptr": 0.0,
793
+ "predicted_margin": 0.08560334369211194,
794
+ "proposal_oracle_success": 0.0,
795
+ "proposal_oracle_utility": 0.07014383375644684,
796
+ "selected_candidate_index": 7.0,
797
+ "selected_success": 0.0,
798
+ "selected_success_gain_over_base": 0.0,
799
+ "selected_utility": 0.5839591026306152,
800
+ "selected_utility_gain_over_base": 0.0,
801
+ "selector_regret": 0.0,
802
+ "success_selector_gap": 0.0,
803
+ "success_support_gap": 1.0,
804
+ "support_gap": 1.9298561662435532,
805
+ "tau": 0.29031104002937436
806
+ },
807
+ "1522272686/train2": {
808
+ "base_success": 0.0,
809
+ "base_utility": 0.5839591026306152,
810
+ "coverage": 0.0,
811
+ "execute_generated": 0.0,
812
+ "fallback_rate": 1.0,
813
+ "hidden_chart_oracle_success": 1.0,
814
+ "hidden_chart_oracle_utility": 2.0,
815
+ "outcome_ptr": 0.0,
816
+ "predicted_margin": 0.07057760540337482,
817
+ "proposal_oracle_success": 0.0,
818
+ "proposal_oracle_utility": 0.14937306940555573,
819
+ "selected_candidate_index": 8.0,
820
+ "selected_success": 0.0,
821
+ "selected_success_gain_over_base": 0.0,
822
+ "selected_utility": 0.5839591026306152,
823
+ "selected_utility_gain_over_base": 0.0,
824
+ "selector_regret": 0.0,
825
+ "success_selector_gap": 0.0,
826
+ "success_support_gap": 1.0,
827
+ "support_gap": 1.8506269305944443,
828
+ "tau": 0.29031104002937436
829
+ },
830
+ "1549324805/train0": {
831
+ "base_success": 0.0,
832
+ "base_utility": 0.10451483726501465,
833
+ "coverage": 1.0,
834
+ "execute_generated": 1.0,
835
+ "fallback_rate": 0.0,
836
+ "hidden_chart_oracle_success": 0.0,
837
+ "hidden_chart_oracle_utility": 0.2974560856819153,
838
+ "outcome_ptr": 0.0,
839
+ "predicted_margin": 0.7033632660299286,
840
+ "proposal_oracle_success": 0.0,
841
+ "proposal_oracle_utility": 0.060871422290802,
842
+ "selected_candidate_index": 0.0,
843
+ "selected_success": 0.0,
844
+ "selected_success_gain_over_base": 0.0,
845
+ "selected_utility": 0.0260902252048254,
846
+ "selected_utility_gain_over_base": -0.07842461206018925,
847
+ "selector_regret": 0.0347811970859766,
848
+ "success_selector_gap": 0.0,
849
+ "success_support_gap": 0.0,
850
+ "support_gap": 0.23658466339111328,
851
+ "tau": 0.29031104002937436
852
+ },
853
+ "1549324805/train1": {
854
+ "base_success": 0.0,
855
+ "base_utility": 0.10451483726501465,
856
+ "coverage": 1.0,
857
+ "execute_generated": 1.0,
858
+ "fallback_rate": 0.0,
859
+ "hidden_chart_oracle_success": 0.0,
860
+ "hidden_chart_oracle_utility": 0.2974560856819153,
861
+ "outcome_ptr": 1.0,
862
+ "predicted_margin": 0.48055562020716125,
863
+ "proposal_oracle_success": 0.0,
864
+ "proposal_oracle_utility": 0.19892151653766632,
865
+ "selected_candidate_index": 0.0,
866
+ "selected_success": 0.0,
867
+ "selected_success_gain_over_base": 0.0,
868
+ "selected_utility": 0.0303483996540308,
869
+ "selected_utility_gain_over_base": -0.07416643761098385,
870
+ "selector_regret": 0.16857311688363552,
871
+ "success_selector_gap": 0.0,
872
+ "success_support_gap": 0.0,
873
+ "support_gap": 0.09853456914424896,
874
+ "tau": 0.29031104002937436
875
+ },
876
+ "1549324805/train2": {
877
+ "base_success": 0.0,
878
+ "base_utility": 0.10451483726501465,
879
+ "coverage": 1.0,
880
+ "execute_generated": 1.0,
881
+ "fallback_rate": 0.0,
882
+ "hidden_chart_oracle_success": 0.0,
883
+ "hidden_chart_oracle_utility": 0.2974560856819153,
884
+ "outcome_ptr": 1.0,
885
+ "predicted_margin": 0.704480055086217,
886
+ "proposal_oracle_success": 0.0,
887
+ "proposal_oracle_utility": 0.11310184001922607,
888
+ "selected_candidate_index": 2.0,
889
+ "selected_success": 0.0,
890
+ "selected_success_gain_over_base": 0.0,
891
+ "selected_utility": 0.03767126798629761,
892
+ "selected_utility_gain_over_base": -0.06684356927871704,
893
+ "selector_regret": 0.07543057203292847,
894
+ "success_selector_gap": 0.0,
895
+ "success_support_gap": 0.0,
896
+ "support_gap": 0.1843542456626892,
897
+ "tau": 0.29031104002937436
898
+ },
899
+ "1578990937/train0": {
900
+ "base_success": 0.0,
901
+ "base_utility": 0.82413649559021,
902
+ "coverage": 0.0,
903
+ "execute_generated": 0.0,
904
+ "fallback_rate": 1.0,
905
+ "hidden_chart_oracle_success": 1.0,
906
+ "hidden_chart_oracle_utility": 2.0,
907
+ "outcome_ptr": 1.0,
908
+ "predicted_margin": 0.2150808050278072,
909
+ "proposal_oracle_success": 1.0,
910
+ "proposal_oracle_utility": 2.0,
911
+ "selected_candidate_index": 6.0,
912
+ "selected_success": 0.0,
913
+ "selected_success_gain_over_base": 0.0,
914
+ "selected_utility": 0.82413649559021,
915
+ "selected_utility_gain_over_base": 0.0,
916
+ "selector_regret": 1.17586350440979,
917
+ "success_selector_gap": 1.0,
918
+ "success_support_gap": 0.0,
919
+ "support_gap": 0.0,
920
+ "tau": 0.29031104002937436
921
+ },
922
+ "1578990937/train1": {
923
+ "base_success": 0.0,
924
+ "base_utility": 0.82413649559021,
925
+ "coverage": 0.0,
926
+ "execute_generated": 0.0,
927
+ "fallback_rate": 1.0,
928
+ "hidden_chart_oracle_success": 1.0,
929
+ "hidden_chart_oracle_utility": 2.0,
930
+ "outcome_ptr": 1.0,
931
+ "predicted_margin": 0.23575927877927283,
932
+ "proposal_oracle_success": 1.0,
933
+ "proposal_oracle_utility": 2.0,
934
+ "selected_candidate_index": 12.0,
935
+ "selected_success": 0.0,
936
+ "selected_success_gain_over_base": 0.0,
937
+ "selected_utility": 0.82413649559021,
938
+ "selected_utility_gain_over_base": 0.0,
939
+ "selector_regret": 1.17586350440979,
940
+ "success_selector_gap": 1.0,
941
+ "success_support_gap": 0.0,
942
+ "support_gap": 0.0,
943
+ "tau": 0.29031104002937436
944
+ },
945
+ "1578990937/train2": {
946
+ "base_success": 0.0,
947
+ "base_utility": 0.82413649559021,
948
+ "coverage": 0.0,
949
+ "execute_generated": 0.0,
950
+ "fallback_rate": 1.0,
951
+ "hidden_chart_oracle_success": 1.0,
952
+ "hidden_chart_oracle_utility": 2.0,
953
+ "outcome_ptr": 1.0,
954
+ "predicted_margin": 0.22754916987219517,
955
+ "proposal_oracle_success": 1.0,
956
+ "proposal_oracle_utility": 2.0,
957
+ "selected_candidate_index": 0.0,
958
+ "selected_success": 0.0,
959
+ "selected_success_gain_over_base": 0.0,
960
+ "selected_utility": 0.82413649559021,
961
+ "selected_utility_gain_over_base": 0.0,
962
+ "selector_regret": 1.17586350440979,
963
+ "success_selector_gap": 1.0,
964
+ "success_support_gap": 0.0,
965
+ "support_gap": 0.0,
966
+ "tau": 0.29031104002937436
967
+ },
968
+ "1747178030/train0": {
969
+ "base_success": 1.0,
970
+ "base_utility": 2.0,
971
+ "coverage": 1.0,
972
+ "execute_generated": 1.0,
973
+ "fallback_rate": 0.0,
974
+ "hidden_chart_oracle_success": 1.0,
975
+ "hidden_chart_oracle_utility": 2.0,
976
+ "outcome_ptr": 0.0,
977
+ "predicted_margin": 0.5849835367136861,
978
+ "proposal_oracle_success": 1.0,
979
+ "proposal_oracle_utility": 2.0,
980
+ "selected_candidate_index": 4.0,
981
+ "selected_success": 1.0,
982
+ "selected_success_gain_over_base": 0.0,
983
+ "selected_utility": 2.0,
984
+ "selected_utility_gain_over_base": 0.0,
985
+ "selector_regret": 0.0,
986
+ "success_selector_gap": 0.0,
987
+ "success_support_gap": 0.0,
988
+ "support_gap": 0.0,
989
+ "tau": 0.29031104002937436
990
+ },
991
+ "1747178030/train1": {
992
+ "base_success": 1.0,
993
+ "base_utility": 2.0,
994
+ "coverage": 1.0,
995
+ "execute_generated": 1.0,
996
+ "fallback_rate": 0.0,
997
+ "hidden_chart_oracle_success": 1.0,
998
+ "hidden_chart_oracle_utility": 2.0,
999
+ "outcome_ptr": 0.0,
1000
+ "predicted_margin": 0.5240784904076439,
1001
+ "proposal_oracle_success": 1.0,
1002
+ "proposal_oracle_utility": 2.0,
1003
+ "selected_candidate_index": 1.0,
1004
+ "selected_success": 1.0,
1005
+ "selected_success_gain_over_base": 0.0,
1006
+ "selected_utility": 2.0,
1007
+ "selected_utility_gain_over_base": 0.0,
1008
+ "selector_regret": 0.0,
1009
+ "success_selector_gap": 0.0,
1010
+ "success_support_gap": 0.0,
1011
+ "support_gap": 0.0,
1012
+ "tau": 0.29031104002937436
1013
+ },
1014
+ "1747178030/train2": {
1015
+ "base_success": 1.0,
1016
+ "base_utility": 2.0,
1017
+ "coverage": 1.0,
1018
+ "execute_generated": 1.0,
1019
+ "fallback_rate": 0.0,
1020
+ "hidden_chart_oracle_success": 1.0,
1021
+ "hidden_chart_oracle_utility": 2.0,
1022
+ "outcome_ptr": 0.0,
1023
+ "predicted_margin": 0.502516346529802,
1024
+ "proposal_oracle_success": 1.0,
1025
+ "proposal_oracle_utility": 2.0,
1026
+ "selected_candidate_index": 0.0,
1027
+ "selected_success": 1.0,
1028
+ "selected_success_gain_over_base": 0.0,
1029
+ "selected_utility": 2.0,
1030
+ "selected_utility_gain_over_base": 0.0,
1031
+ "selector_regret": 0.0,
1032
+ "success_selector_gap": 0.0,
1033
+ "success_support_gap": 0.0,
1034
+ "support_gap": 0.0,
1035
+ "tau": 0.29031104002937436
1036
+ },
1037
+ "1790188787/train0": {
1038
+ "base_success": 0.0,
1039
+ "base_utility": 0.12476746737957001,
1040
+ "coverage": 1.0,
1041
+ "execute_generated": 1.0,
1042
+ "fallback_rate": 0.0,
1043
+ "hidden_chart_oracle_success": 1.0,
1044
+ "hidden_chart_oracle_utility": 2.0,
1045
+ "outcome_ptr": 1.0,
1046
+ "predicted_margin": 0.7041875516262702,
1047
+ "proposal_oracle_success": 1.0,
1048
+ "proposal_oracle_utility": 2.0,
1049
+ "selected_candidate_index": 5.0,
1050
+ "selected_success": 1.0,
1051
+ "selected_success_gain_over_base": 1.0,
1052
+ "selected_utility": 2.0,
1053
+ "selected_utility_gain_over_base": 1.87523253262043,
1054
+ "selector_regret": 0.0,
1055
+ "success_selector_gap": 0.0,
1056
+ "success_support_gap": 0.0,
1057
+ "support_gap": 0.0,
1058
+ "tau": 0.29031104002937436
1059
+ },
1060
+ "1790188787/train1": {
1061
+ "base_success": 0.0,
1062
+ "base_utility": 0.12476746737957001,
1063
+ "coverage": 1.0,
1064
+ "execute_generated": 1.0,
1065
+ "fallback_rate": 0.0,
1066
+ "hidden_chart_oracle_success": 1.0,
1067
+ "hidden_chart_oracle_utility": 2.0,
1068
+ "outcome_ptr": 1.0,
1069
+ "predicted_margin": 0.719052615383995,
1070
+ "proposal_oracle_success": 1.0,
1071
+ "proposal_oracle_utility": 2.0,
1072
+ "selected_candidate_index": 6.0,
1073
+ "selected_success": 0.0,
1074
+ "selected_success_gain_over_base": 0.0,
1075
+ "selected_utility": 0.13741976022720337,
1076
+ "selected_utility_gain_over_base": 0.012652292847633362,
1077
+ "selector_regret": 1.8625802397727966,
1078
+ "success_selector_gap": 1.0,
1079
+ "success_support_gap": 0.0,
1080
+ "support_gap": 0.0,
1081
+ "tau": 0.29031104002937436
1082
+ },
1083
+ "1790188787/train2": {
1084
+ "base_success": 0.0,
1085
+ "base_utility": 0.12476746737957001,
1086
+ "coverage": 1.0,
1087
+ "execute_generated": 1.0,
1088
+ "fallback_rate": 0.0,
1089
+ "hidden_chart_oracle_success": 1.0,
1090
+ "hidden_chart_oracle_utility": 2.0,
1091
+ "outcome_ptr": 1.0,
1092
+ "predicted_margin": 0.719052615383995,
1093
+ "proposal_oracle_success": 1.0,
1094
+ "proposal_oracle_utility": 2.0,
1095
+ "selected_candidate_index": 8.0,
1096
+ "selected_success": 0.0,
1097
+ "selected_success_gain_over_base": 0.0,
1098
+ "selected_utility": 0.13743820786476135,
1099
+ "selected_utility_gain_over_base": 0.012670740485191345,
1100
+ "selector_regret": 1.8625617921352386,
1101
+ "success_selector_gap": 1.0,
1102
+ "success_support_gap": 0.0,
1103
+ "support_gap": 0.0,
1104
+ "tau": 0.29031104002937436
1105
+ },
1106
+ "1822210471/train0": {
1107
+ "base_success": 0.0,
1108
+ "base_utility": 0.07191530615091324,
1109
+ "coverage": 0.0,
1110
+ "execute_generated": 0.0,
1111
+ "fallback_rate": 1.0,
1112
+ "hidden_chart_oracle_success": 1.0,
1113
+ "hidden_chart_oracle_utility": 2.0,
1114
+ "outcome_ptr": 1.0,
1115
+ "predicted_margin": 0.049984085020529354,
1116
+ "proposal_oracle_success": 1.0,
1117
+ "proposal_oracle_utility": 2.0,
1118
+ "selected_candidate_index": 4.0,
1119
+ "selected_success": 0.0,
1120
+ "selected_success_gain_over_base": 0.0,
1121
+ "selected_utility": 0.07191530615091324,
1122
+ "selected_utility_gain_over_base": 0.0,
1123
+ "selector_regret": 1.9280846938490868,
1124
+ "success_selector_gap": 1.0,
1125
+ "success_support_gap": 0.0,
1126
+ "support_gap": 0.0,
1127
+ "tau": 0.29031104002937436
1128
+ },
1129
+ "1822210471/train1": {
1130
+ "base_success": 0.0,
1131
+ "base_utility": 0.07191530615091324,
1132
+ "coverage": 0.0,
1133
+ "execute_generated": 0.0,
1134
+ "fallback_rate": 1.0,
1135
+ "hidden_chart_oracle_success": 1.0,
1136
+ "hidden_chart_oracle_utility": 2.0,
1137
+ "outcome_ptr": 1.0,
1138
+ "predicted_margin": 0.049984085020529354,
1139
+ "proposal_oracle_success": 1.0,
1140
+ "proposal_oracle_utility": 2.0,
1141
+ "selected_candidate_index": 4.0,
1142
+ "selected_success": 0.0,
1143
+ "selected_success_gain_over_base": 0.0,
1144
+ "selected_utility": 0.07191530615091324,
1145
+ "selected_utility_gain_over_base": 0.0,
1146
+ "selector_regret": 1.9280846938490868,
1147
+ "success_selector_gap": 1.0,
1148
+ "success_support_gap": 0.0,
1149
+ "support_gap": 0.0,
1150
+ "tau": 0.29031104002937436
1151
+ },
1152
+ "1822210471/train2": {
1153
+ "base_success": 0.0,
1154
+ "base_utility": 0.07191530615091324,
1155
+ "coverage": 0.0,
1156
+ "execute_generated": 0.0,
1157
+ "fallback_rate": 1.0,
1158
+ "hidden_chart_oracle_success": 1.0,
1159
+ "hidden_chart_oracle_utility": 2.0,
1160
+ "outcome_ptr": 1.0,
1161
+ "predicted_margin": 0.049984085020529354,
1162
+ "proposal_oracle_success": 1.0,
1163
+ "proposal_oracle_utility": 2.0,
1164
+ "selected_candidate_index": 10.0,
1165
+ "selected_success": 0.0,
1166
+ "selected_success_gain_over_base": 0.0,
1167
+ "selected_utility": 0.07191530615091324,
1168
+ "selected_utility_gain_over_base": 0.0,
1169
+ "selector_regret": 1.9280846938490868,
1170
+ "success_selector_gap": 1.0,
1171
+ "success_support_gap": 0.0,
1172
+ "support_gap": 0.0,
1173
+ "tau": 0.29031104002937436
1174
+ },
1175
+ "1845489013/train0": {
1176
+ "base_success": 1.0,
1177
+ "base_utility": 2.0,
1178
+ "coverage": 0.0,
1179
+ "execute_generated": 0.0,
1180
+ "fallback_rate": 1.0,
1181
+ "hidden_chart_oracle_success": 1.0,
1182
+ "hidden_chart_oracle_utility": 2.0,
1183
+ "outcome_ptr": 0.0,
1184
+ "predicted_margin": 0.24672540504885201,
1185
+ "proposal_oracle_success": 1.0,
1186
+ "proposal_oracle_utility": 2.0,
1187
+ "selected_candidate_index": 10.0,
1188
+ "selected_success": 1.0,
1189
+ "selected_success_gain_over_base": 0.0,
1190
+ "selected_utility": 2.0,
1191
+ "selected_utility_gain_over_base": 0.0,
1192
+ "selector_regret": 0.0,
1193
+ "success_selector_gap": 0.0,
1194
+ "success_support_gap": 0.0,
1195
+ "support_gap": 0.0,
1196
+ "tau": 0.29031104002937436
1197
+ },
1198
+ "1845489013/train1": {
1199
+ "base_success": 1.0,
1200
+ "base_utility": 2.0,
1201
+ "coverage": 1.0,
1202
+ "execute_generated": 1.0,
1203
+ "fallback_rate": 0.0,
1204
+ "hidden_chart_oracle_success": 1.0,
1205
+ "hidden_chart_oracle_utility": 2.0,
1206
+ "outcome_ptr": 0.0,
1207
+ "predicted_margin": 0.2978540613458687,
1208
+ "proposal_oracle_success": 1.0,
1209
+ "proposal_oracle_utility": 2.0,
1210
+ "selected_candidate_index": 3.0,
1211
+ "selected_success": 1.0,
1212
+ "selected_success_gain_over_base": 0.0,
1213
+ "selected_utility": 2.0,
1214
+ "selected_utility_gain_over_base": 0.0,
1215
+ "selector_regret": 0.0,
1216
+ "success_selector_gap": 0.0,
1217
+ "success_support_gap": 0.0,
1218
+ "support_gap": 0.0,
1219
+ "tau": 0.29031104002937436
1220
+ },
1221
+ "1845489013/train2": {
1222
+ "base_success": 1.0,
1223
+ "base_utility": 2.0,
1224
+ "coverage": 1.0,
1225
+ "execute_generated": 1.0,
1226
+ "fallback_rate": 0.0,
1227
+ "hidden_chart_oracle_success": 1.0,
1228
+ "hidden_chart_oracle_utility": 2.0,
1229
+ "outcome_ptr": 0.0,
1230
+ "predicted_margin": 0.2978540613458687,
1231
+ "proposal_oracle_success": 1.0,
1232
+ "proposal_oracle_utility": 2.0,
1233
+ "selected_candidate_index": 11.0,
1234
+ "selected_success": 1.0,
1235
+ "selected_success_gain_over_base": 0.0,
1236
+ "selected_utility": 2.0,
1237
+ "selected_utility_gain_over_base": 0.0,
1238
+ "selector_regret": 0.0,
1239
+ "success_selector_gap": 0.0,
1240
+ "success_support_gap": 0.0,
1241
+ "support_gap": 0.0,
1242
+ "tau": 0.29031104002937436
1243
+ },
1244
+ "1999022574/train0": {
1245
+ "base_success": 1.0,
1246
+ "base_utility": 2.0,
1247
+ "coverage": 0.0,
1248
+ "execute_generated": 0.0,
1249
+ "fallback_rate": 1.0,
1250
+ "hidden_chart_oracle_success": 1.0,
1251
+ "hidden_chart_oracle_utility": 2.0,
1252
+ "outcome_ptr": 0.0,
1253
+ "predicted_margin": 0.2755208385556546,
1254
+ "proposal_oracle_success": 1.0,
1255
+ "proposal_oracle_utility": 2.0,
1256
+ "selected_candidate_index": 5.0,
1257
+ "selected_success": 1.0,
1258
+ "selected_success_gain_over_base": 0.0,
1259
+ "selected_utility": 2.0,
1260
+ "selected_utility_gain_over_base": 0.0,
1261
+ "selector_regret": 0.0,
1262
+ "success_selector_gap": 0.0,
1263
+ "success_support_gap": 0.0,
1264
+ "support_gap": 0.0,
1265
+ "tau": 0.29031104002937436
1266
+ },
1267
+ "1999022574/train1": {
1268
+ "base_success": 1.0,
1269
+ "base_utility": 2.0,
1270
+ "coverage": 1.0,
1271
+ "execute_generated": 1.0,
1272
+ "fallback_rate": 0.0,
1273
+ "hidden_chart_oracle_success": 1.0,
1274
+ "hidden_chart_oracle_utility": 2.0,
1275
+ "outcome_ptr": 0.0,
1276
+ "predicted_margin": 0.3312653260052494,
1277
+ "proposal_oracle_success": 1.0,
1278
+ "proposal_oracle_utility": 2.0,
1279
+ "selected_candidate_index": 9.0,
1280
+ "selected_success": 0.0,
1281
+ "selected_success_gain_over_base": -1.0,
1282
+ "selected_utility": 0.8125,
1283
+ "selected_utility_gain_over_base": -1.1875,
1284
+ "selector_regret": 1.1875,
1285
+ "success_selector_gap": 1.0,
1286
+ "success_support_gap": 0.0,
1287
+ "support_gap": 0.0,
1288
+ "tau": 0.29031104002937436
1289
+ },
1290
+ "1999022574/train2": {
1291
+ "base_success": 1.0,
1292
+ "base_utility": 2.0,
1293
+ "coverage": 1.0,
1294
+ "execute_generated": 1.0,
1295
+ "fallback_rate": 0.0,
1296
+ "hidden_chart_oracle_success": 1.0,
1297
+ "hidden_chart_oracle_utility": 2.0,
1298
+ "outcome_ptr": 0.0,
1299
+ "predicted_margin": 0.3176403762902519,
1300
+ "proposal_oracle_success": 1.0,
1301
+ "proposal_oracle_utility": 2.0,
1302
+ "selected_candidate_index": 6.0,
1303
+ "selected_success": 0.0,
1304
+ "selected_success_gain_over_base": -1.0,
1305
+ "selected_utility": 0.1785425841808319,
1306
+ "selected_utility_gain_over_base": -1.821457415819168,
1307
+ "selector_regret": 1.821457415819168,
1308
+ "success_selector_gap": 1.0,
1309
+ "success_support_gap": 0.0,
1310
+ "support_gap": 0.0,
1311
+ "tau": 0.29031104002937436
1312
+ },
1313
+ "2012211287/train0": {
1314
+ "base_success": 0.0,
1315
+ "base_utility": 0.7236053347587585,
1316
+ "coverage": 0.0,
1317
+ "execute_generated": 0.0,
1318
+ "fallback_rate": 1.0,
1319
+ "hidden_chart_oracle_success": 1.0,
1320
+ "hidden_chart_oracle_utility": 2.0,
1321
+ "outcome_ptr": 1.0,
1322
+ "predicted_margin": 0.17828726626426614,
1323
+ "proposal_oracle_success": 1.0,
1324
+ "proposal_oracle_utility": 2.0,
1325
+ "selected_candidate_index": 10.0,
1326
+ "selected_success": 0.0,
1327
+ "selected_success_gain_over_base": 0.0,
1328
+ "selected_utility": 0.7236053347587585,
1329
+ "selected_utility_gain_over_base": 0.0,
1330
+ "selector_regret": 1.2763946652412415,
1331
+ "success_selector_gap": 1.0,
1332
+ "success_support_gap": 0.0,
1333
+ "support_gap": 0.0,
1334
+ "tau": 0.29031104002937436
1335
+ },
1336
+ "2012211287/train1": {
1337
+ "base_success": 0.0,
1338
+ "base_utility": 0.7236053347587585,
1339
+ "coverage": 0.0,
1340
+ "execute_generated": 0.0,
1341
+ "fallback_rate": 1.0,
1342
+ "hidden_chart_oracle_success": 1.0,
1343
+ "hidden_chart_oracle_utility": 2.0,
1344
+ "outcome_ptr": 1.0,
1345
+ "predicted_margin": 0.20414937358844587,
1346
+ "proposal_oracle_success": 1.0,
1347
+ "proposal_oracle_utility": 2.0,
1348
+ "selected_candidate_index": 15.0,
1349
+ "selected_success": 0.0,
1350
+ "selected_success_gain_over_base": 0.0,
1351
+ "selected_utility": 0.7236053347587585,
1352
+ "selected_utility_gain_over_base": 0.0,
1353
+ "selector_regret": 1.2763946652412415,
1354
+ "success_selector_gap": 1.0,
1355
+ "success_support_gap": 0.0,
1356
+ "support_gap": 0.0,
1357
+ "tau": 0.29031104002937436
1358
+ },
1359
+ "2012211287/train2": {
1360
+ "base_success": 0.0,
1361
+ "base_utility": 0.7236053347587585,
1362
+ "coverage": 0.0,
1363
+ "execute_generated": 0.0,
1364
+ "fallback_rate": 1.0,
1365
+ "hidden_chart_oracle_success": 1.0,
1366
+ "hidden_chart_oracle_utility": 2.0,
1367
+ "outcome_ptr": 1.0,
1368
+ "predicted_margin": 0.1978378634332485,
1369
+ "proposal_oracle_success": 1.0,
1370
+ "proposal_oracle_utility": 2.0,
1371
+ "selected_candidate_index": 4.0,
1372
+ "selected_success": 0.0,
1373
+ "selected_success_gain_over_base": 0.0,
1374
+ "selected_utility": 0.7236053347587585,
1375
+ "selected_utility_gain_over_base": 0.0,
1376
+ "selector_regret": 1.2763946652412415,
1377
+ "success_selector_gap": 1.0,
1378
+ "success_support_gap": 0.0,
1379
+ "support_gap": 0.0,
1380
+ "tau": 0.29031104002937436
1381
+ },
1382
+ "2038594191/train0": {
1383
+ "base_success": 1.0,
1384
+ "base_utility": 2.0,
1385
+ "coverage": 1.0,
1386
+ "execute_generated": 1.0,
1387
+ "fallback_rate": 0.0,
1388
+ "hidden_chart_oracle_success": 1.0,
1389
+ "hidden_chart_oracle_utility": 2.0,
1390
+ "outcome_ptr": 0.0,
1391
+ "predicted_margin": 0.3359117235345434,
1392
+ "proposal_oracle_success": 1.0,
1393
+ "proposal_oracle_utility": 2.0,
1394
+ "selected_candidate_index": 0.0,
1395
+ "selected_success": 1.0,
1396
+ "selected_success_gain_over_base": 0.0,
1397
+ "selected_utility": 2.0,
1398
+ "selected_utility_gain_over_base": 0.0,
1399
+ "selector_regret": 0.0,
1400
+ "success_selector_gap": 0.0,
1401
+ "success_support_gap": 0.0,
1402
+ "support_gap": 0.0,
1403
+ "tau": 0.29031104002937436
1404
+ },
1405
+ "2038594191/train1": {
1406
+ "base_success": 1.0,
1407
+ "base_utility": 2.0,
1408
+ "coverage": 1.0,
1409
+ "execute_generated": 1.0,
1410
+ "fallback_rate": 0.0,
1411
+ "hidden_chart_oracle_success": 1.0,
1412
+ "hidden_chart_oracle_utility": 2.0,
1413
+ "outcome_ptr": 0.0,
1414
+ "predicted_margin": 0.37949252862164845,
1415
+ "proposal_oracle_success": 1.0,
1416
+ "proposal_oracle_utility": 2.0,
1417
+ "selected_candidate_index": 4.0,
1418
+ "selected_success": 1.0,
1419
+ "selected_success_gain_over_base": 0.0,
1420
+ "selected_utility": 2.0,
1421
+ "selected_utility_gain_over_base": 0.0,
1422
+ "selector_regret": 0.0,
1423
+ "success_selector_gap": 0.0,
1424
+ "success_support_gap": 0.0,
1425
+ "support_gap": 0.0,
1426
+ "tau": 0.29031104002937436
1427
+ },
1428
+ "2038594191/train2": {
1429
+ "base_success": 1.0,
1430
+ "base_utility": 2.0,
1431
+ "coverage": 1.0,
1432
+ "execute_generated": 1.0,
1433
+ "fallback_rate": 0.0,
1434
+ "hidden_chart_oracle_success": 1.0,
1435
+ "hidden_chart_oracle_utility": 2.0,
1436
+ "outcome_ptr": 0.0,
1437
+ "predicted_margin": 0.3466799479100969,
1438
+ "proposal_oracle_success": 1.0,
1439
+ "proposal_oracle_utility": 2.0,
1440
+ "selected_candidate_index": 1.0,
1441
+ "selected_success": 1.0,
1442
+ "selected_success_gain_over_base": 0.0,
1443
+ "selected_utility": 2.0,
1444
+ "selected_utility_gain_over_base": 0.0,
1445
+ "selector_regret": 0.0,
1446
+ "success_selector_gap": 0.0,
1447
+ "success_support_gap": 0.0,
1448
+ "support_gap": 0.0,
1449
+ "tau": 0.29031104002937436
1450
+ },
1451
+ "2054899126/train0": {
1452
+ "base_success": 1.0,
1453
+ "base_utility": 2.0,
1454
+ "coverage": 1.0,
1455
+ "execute_generated": 1.0,
1456
+ "fallback_rate": 0.0,
1457
+ "hidden_chart_oracle_success": 1.0,
1458
+ "hidden_chart_oracle_utility": 2.0,
1459
+ "outcome_ptr": 0.0,
1460
+ "predicted_margin": 0.8015567150083238,
1461
+ "proposal_oracle_success": 0.5,
1462
+ "proposal_oracle_utility": 1.077283151447773,
1463
+ "selected_candidate_index": 0.0,
1464
+ "selected_success": 0.5,
1465
+ "selected_success_gain_over_base": -0.5,
1466
+ "selected_utility": 1.052448581904173,
1467
+ "selected_utility_gain_over_base": -0.9475514180958271,
1468
+ "selector_regret": 0.024834569543600082,
1469
+ "success_selector_gap": 0.0,
1470
+ "success_support_gap": 0.5,
1471
+ "support_gap": 0.922716848552227,
1472
+ "tau": 0.29031104002937436
1473
+ },
1474
+ "2054899126/train1": {
1475
+ "base_success": 1.0,
1476
+ "base_utility": 2.0,
1477
+ "coverage": 1.0,
1478
+ "execute_generated": 1.0,
1479
+ "fallback_rate": 0.0,
1480
+ "hidden_chart_oracle_success": 1.0,
1481
+ "hidden_chart_oracle_utility": 2.0,
1482
+ "outcome_ptr": 0.0,
1483
+ "predicted_margin": 0.7535063374357802,
1484
+ "proposal_oracle_success": 1.0,
1485
+ "proposal_oracle_utility": 2.0,
1486
+ "selected_candidate_index": 0.5,
1487
+ "selected_success": 0.5,
1488
+ "selected_success_gain_over_base": -0.5,
1489
+ "selected_utility": 1.0664558708667755,
1490
+ "selected_utility_gain_over_base": -0.9335441291332245,
1491
+ "selector_regret": 0.9335441291332245,
1492
+ "success_selector_gap": 0.5,
1493
+ "success_support_gap": 0.0,
1494
+ "support_gap": 0.0,
1495
+ "tau": 0.29031104002937436
1496
+ },
1497
+ "2054899126/train2": {
1498
+ "base_success": 1.0,
1499
+ "base_utility": 2.0,
1500
+ "coverage": 1.0,
1501
+ "execute_generated": 1.0,
1502
+ "fallback_rate": 0.0,
1503
+ "hidden_chart_oracle_success": 1.0,
1504
+ "hidden_chart_oracle_utility": 2.0,
1505
+ "outcome_ptr": 0.0,
1506
+ "predicted_margin": 0.7376074795110784,
1507
+ "proposal_oracle_success": 1.0,
1508
+ "proposal_oracle_utility": 2.0,
1509
+ "selected_candidate_index": 3.5,
1510
+ "selected_success": 0.5,
1511
+ "selected_success_gain_over_base": -0.5,
1512
+ "selected_utility": 1.0326566398143768,
1513
+ "selected_utility_gain_over_base": -0.9673433601856232,
1514
+ "selector_regret": 0.9673433601856232,
1515
+ "success_selector_gap": 0.5,
1516
+ "success_support_gap": 0.0,
1517
+ "support_gap": 0.0,
1518
+ "tau": 0.29031104002937436
1519
+ },
1520
+ "2061635888/train0": {
1521
+ "base_success": 0.0,
1522
+ "base_utility": 0.05225050449371338,
1523
+ "coverage": 0.0,
1524
+ "execute_generated": 0.0,
1525
+ "fallback_rate": 1.0,
1526
+ "hidden_chart_oracle_success": 1.0,
1527
+ "hidden_chart_oracle_utility": 2.0,
1528
+ "outcome_ptr": 1.0,
1529
+ "predicted_margin": 0.13472633878966278,
1530
+ "proposal_oracle_success": 0.0,
1531
+ "proposal_oracle_utility": 0.09799930453300476,
1532
+ "selected_candidate_index": 5.0,
1533
+ "selected_success": 0.0,
1534
+ "selected_success_gain_over_base": 0.0,
1535
+ "selected_utility": 0.05225050449371338,
1536
+ "selected_utility_gain_over_base": 0.0,
1537
+ "selector_regret": 0.04574880003929138,
1538
+ "success_selector_gap": 0.0,
1539
+ "success_support_gap": 1.0,
1540
+ "support_gap": 1.9020006954669952,
1541
+ "tau": 0.29031104002937436
1542
+ },
1543
+ "2061635888/train1": {
1544
+ "base_success": 0.0,
1545
+ "base_utility": 0.05225050449371338,
1546
+ "coverage": 0.0,
1547
+ "execute_generated": 0.0,
1548
+ "fallback_rate": 1.0,
1549
+ "hidden_chart_oracle_success": 1.0,
1550
+ "hidden_chart_oracle_utility": 2.0,
1551
+ "outcome_ptr": 1.0,
1552
+ "predicted_margin": 0.14234590579614123,
1553
+ "proposal_oracle_success": 0.0,
1554
+ "proposal_oracle_utility": 0.07580336928367615,
1555
+ "selected_candidate_index": 8.0,
1556
+ "selected_success": 0.0,
1557
+ "selected_success_gain_over_base": 0.0,
1558
+ "selected_utility": 0.05225050449371338,
1559
+ "selected_utility_gain_over_base": 0.0,
1560
+ "selector_regret": 0.02355286478996277,
1561
+ "success_selector_gap": 0.0,
1562
+ "success_support_gap": 1.0,
1563
+ "support_gap": 1.9241966307163239,
1564
+ "tau": 0.29031104002937436
1565
+ },
1566
+ "2061635888/train2": {
1567
+ "base_success": 0.0,
1568
+ "base_utility": 0.05225050449371338,
1569
+ "coverage": 0.0,
1570
+ "execute_generated": 0.0,
1571
+ "fallback_rate": 1.0,
1572
+ "hidden_chart_oracle_success": 1.0,
1573
+ "hidden_chart_oracle_utility": 2.0,
1574
+ "outcome_ptr": 1.0,
1575
+ "predicted_margin": 0.13062938104776378,
1576
+ "proposal_oracle_success": 0.0,
1577
+ "proposal_oracle_utility": 0.11105132102966309,
1578
+ "selected_candidate_index": 5.0,
1579
+ "selected_success": 0.0,
1580
+ "selected_success_gain_over_base": 0.0,
1581
+ "selected_utility": 0.05225050449371338,
1582
+ "selected_utility_gain_over_base": 0.0,
1583
+ "selector_regret": 0.05880081653594971,
1584
+ "success_selector_gap": 0.0,
1585
+ "success_support_gap": 1.0,
1586
+ "support_gap": 1.888948678970337,
1587
+ "tau": 0.29031104002937436
1588
+ },
1589
+ "211245842/train0": {
1590
+ "base_success": 0.0,
1591
+ "base_utility": 0.19090841710567474,
1592
+ "coverage": 1.0,
1593
+ "execute_generated": 1.0,
1594
+ "fallback_rate": 0.0,
1595
+ "hidden_chart_oracle_success": 1.0,
1596
+ "hidden_chart_oracle_utility": 2.0,
1597
+ "outcome_ptr": 1.0,
1598
+ "predicted_margin": 0.3392426447136978,
1599
+ "proposal_oracle_success": 1.0,
1600
+ "proposal_oracle_utility": 2.0,
1601
+ "selected_candidate_index": 11.0,
1602
+ "selected_success": 1.0,
1603
+ "selected_success_gain_over_base": 1.0,
1604
+ "selected_utility": 2.0,
1605
+ "selected_utility_gain_over_base": 1.8090915828943253,
1606
+ "selector_regret": 0.0,
1607
+ "success_selector_gap": 0.0,
1608
+ "success_support_gap": 0.0,
1609
+ "support_gap": 0.0,
1610
+ "tau": 0.29031104002937436
1611
+ },
1612
+ "211245842/train1": {
1613
+ "base_success": 0.0,
1614
+ "base_utility": 0.19090841710567474,
1615
+ "coverage": 1.0,
1616
+ "execute_generated": 1.0,
1617
+ "fallback_rate": 0.0,
1618
+ "hidden_chart_oracle_success": 1.0,
1619
+ "hidden_chart_oracle_utility": 2.0,
1620
+ "outcome_ptr": 1.0,
1621
+ "predicted_margin": 0.3721286611771941,
1622
+ "proposal_oracle_success": 0.0,
1623
+ "proposal_oracle_utility": 0.21295717358589172,
1624
+ "selected_candidate_index": 15.0,
1625
+ "selected_success": 0.0,
1626
+ "selected_success_gain_over_base": 0.0,
1627
+ "selected_utility": 0.14139282703399658,
1628
+ "selected_utility_gain_over_base": -0.04951559007167816,
1629
+ "selector_regret": 0.07156434655189514,
1630
+ "success_selector_gap": 0.0,
1631
+ "success_support_gap": 1.0,
1632
+ "support_gap": 1.7870428264141083,
1633
+ "tau": 0.29031104002937436
1634
+ },
1635
+ "211245842/train2": {
1636
+ "base_success": 0.0,
1637
+ "base_utility": 0.19090841710567474,
1638
+ "coverage": 1.0,
1639
+ "execute_generated": 1.0,
1640
+ "fallback_rate": 0.0,
1641
+ "hidden_chart_oracle_success": 1.0,
1642
+ "hidden_chart_oracle_utility": 2.0,
1643
+ "outcome_ptr": 1.0,
1644
+ "predicted_margin": 0.3899264429033258,
1645
+ "proposal_oracle_success": 0.0,
1646
+ "proposal_oracle_utility": 0.23388276994228363,
1647
+ "selected_candidate_index": 15.0,
1648
+ "selected_success": 0.0,
1649
+ "selected_success_gain_over_base": 0.0,
1650
+ "selected_utility": 0.13791468739509583,
1651
+ "selected_utility_gain_over_base": -0.05299372971057892,
1652
+ "selector_regret": 0.0959680825471878,
1653
+ "success_selector_gap": 0.0,
1654
+ "success_support_gap": 1.0,
1655
+ "support_gap": 1.7661172300577164,
1656
+ "tau": 0.29031104002937436
1657
+ },
1658
+ "278008524/train0": {
1659
+ "base_success": 0.0,
1660
+ "base_utility": 0.042734989896416664,
1661
+ "coverage": 1.0,
1662
+ "execute_generated": 1.0,
1663
+ "fallback_rate": 0.0,
1664
+ "hidden_chart_oracle_success": 0.5,
1665
+ "hidden_chart_oracle_utility": 1.1139086410403252,
1666
+ "outcome_ptr": 1.0,
1667
+ "predicted_margin": 0.48982706754483685,
1668
+ "proposal_oracle_success": 0.5,
1669
+ "proposal_oracle_utility": 1.0733563229441643,
1670
+ "selected_candidate_index": 5.5,
1671
+ "selected_success": 0.0,
1672
+ "selected_success_gain_over_base": 0.0,
1673
+ "selected_utility": 0.0616548378020525,
1674
+ "selected_utility_gain_over_base": 0.018919847905635834,
1675
+ "selector_regret": 1.0117014851421118,
1676
+ "success_selector_gap": 0.5,
1677
+ "success_support_gap": 0.0,
1678
+ "support_gap": 0.04055231809616089,
1679
+ "tau": 0.29031104002937436
1680
+ },
1681
+ "278008524/train1": {
1682
+ "base_success": 0.0,
1683
+ "base_utility": 0.042734989896416664,
1684
+ "coverage": 1.0,
1685
+ "execute_generated": 1.0,
1686
+ "fallback_rate": 0.0,
1687
+ "hidden_chart_oracle_success": 0.5,
1688
+ "hidden_chart_oracle_utility": 1.1139086410403252,
1689
+ "outcome_ptr": 1.0,
1690
+ "predicted_margin": 0.523516952673019,
1691
+ "proposal_oracle_success": 0.5,
1692
+ "proposal_oracle_utility": 1.0376745946705341,
1693
+ "selected_candidate_index": 2.5,
1694
+ "selected_success": 0.0,
1695
+ "selected_success_gain_over_base": 0.0,
1696
+ "selected_utility": 0.06736786291003227,
1697
+ "selected_utility_gain_over_base": 0.024632873013615608,
1698
+ "selector_regret": 0.9703067317605019,
1699
+ "success_selector_gap": 0.5,
1700
+ "success_support_gap": 0.0,
1701
+ "support_gap": 0.07623404636979103,
1702
+ "tau": 0.29031104002937436
1703
+ },
1704
+ "278008524/train2": {
1705
+ "base_success": 0.0,
1706
+ "base_utility": 0.042734989896416664,
1707
+ "coverage": 1.0,
1708
+ "execute_generated": 1.0,
1709
+ "fallback_rate": 0.0,
1710
+ "hidden_chart_oracle_success": 0.5,
1711
+ "hidden_chart_oracle_utility": 1.1139086410403252,
1712
+ "outcome_ptr": 1.0,
1713
+ "predicted_margin": 0.47410396996311766,
1714
+ "proposal_oracle_success": 0.5,
1715
+ "proposal_oracle_utility": 1.1359489113092422,
1716
+ "selected_candidate_index": 9.5,
1717
+ "selected_success": 0.5,
1718
+ "selected_success_gain_over_base": 0.5,
1719
+ "selected_utility": 1.034129787236452,
1720
+ "selected_utility_gain_over_base": 0.9913947973400354,
1721
+ "selector_regret": 0.10181912407279015,
1722
+ "success_selector_gap": 0.0,
1723
+ "success_support_gap": 0.0,
1724
+ "support_gap": 0.0,
1725
+ "tau": 0.29031104002937436
1726
+ },
1727
+ "307068879/train0": {
1728
+ "base_success": 0.0,
1729
+ "base_utility": 0.6000905632972717,
1730
+ "coverage": 0.0,
1731
+ "execute_generated": 0.0,
1732
+ "fallback_rate": 1.0,
1733
+ "hidden_chart_oracle_success": 1.0,
1734
+ "hidden_chart_oracle_utility": 2.0,
1735
+ "outcome_ptr": 0.0,
1736
+ "predicted_margin": 0.28418063141808364,
1737
+ "proposal_oracle_success": 0.0,
1738
+ "proposal_oracle_utility": 0.2115575522184372,
1739
+ "selected_candidate_index": 2.0,
1740
+ "selected_success": 0.0,
1741
+ "selected_success_gain_over_base": 0.0,
1742
+ "selected_utility": 0.6000905632972717,
1743
+ "selected_utility_gain_over_base": 0.0,
1744
+ "selector_regret": 0.0,
1745
+ "success_selector_gap": 0.0,
1746
+ "success_support_gap": 1.0,
1747
+ "support_gap": 1.7884424477815628,
1748
+ "tau": 0.29031104002937436
1749
+ },
1750
+ "307068879/train1": {
1751
+ "base_success": 0.0,
1752
+ "base_utility": 0.6000905632972717,
1753
+ "coverage": 0.0,
1754
+ "execute_generated": 0.0,
1755
+ "fallback_rate": 1.0,
1756
+ "hidden_chart_oracle_success": 1.0,
1757
+ "hidden_chart_oracle_utility": 2.0,
1758
+ "outcome_ptr": 0.0,
1759
+ "predicted_margin": 0.2759242006849786,
1760
+ "proposal_oracle_success": 0.0,
1761
+ "proposal_oracle_utility": 0.5876073241233826,
1762
+ "selected_candidate_index": 7.0,
1763
+ "selected_success": 0.0,
1764
+ "selected_success_gain_over_base": 0.0,
1765
+ "selected_utility": 0.6000905632972717,
1766
+ "selected_utility_gain_over_base": 0.0,
1767
+ "selector_regret": 0.0,
1768
+ "success_selector_gap": 0.0,
1769
+ "success_support_gap": 1.0,
1770
+ "support_gap": 1.4123926758766174,
1771
+ "tau": 0.29031104002937436
1772
+ },
1773
+ "307068879/train2": {
1774
+ "base_success": 0.0,
1775
+ "base_utility": 0.6000905632972717,
1776
+ "coverage": 1.0,
1777
+ "execute_generated": 1.0,
1778
+ "fallback_rate": 0.0,
1779
+ "hidden_chart_oracle_success": 1.0,
1780
+ "hidden_chart_oracle_utility": 2.0,
1781
+ "outcome_ptr": 0.0,
1782
+ "predicted_margin": 0.2917507478363105,
1783
+ "proposal_oracle_success": 0.0,
1784
+ "proposal_oracle_utility": 0.2302134484052658,
1785
+ "selected_candidate_index": 0.0,
1786
+ "selected_success": 0.0,
1787
+ "selected_success_gain_over_base": 0.0,
1788
+ "selected_utility": 0.22513391077518463,
1789
+ "selected_utility_gain_over_base": -0.3749566525220871,
1790
+ "selector_regret": 0.005079537630081177,
1791
+ "success_selector_gap": 0.0,
1792
+ "success_support_gap": 1.0,
1793
+ "support_gap": 1.7697865515947342,
1794
+ "tau": 0.29031104002937436
1795
+ },
1796
+ "412102563/train0": {
1797
+ "base_success": 1.0,
1798
+ "base_utility": 2.0,
1799
+ "coverage": 0.0,
1800
+ "execute_generated": 0.0,
1801
+ "fallback_rate": 1.0,
1802
+ "hidden_chart_oracle_success": 1.0,
1803
+ "hidden_chart_oracle_utility": 2.0,
1804
+ "outcome_ptr": 0.0,
1805
+ "predicted_margin": 0.1966920683523119,
1806
+ "proposal_oracle_success": 1.0,
1807
+ "proposal_oracle_utility": 2.0,
1808
+ "selected_candidate_index": 13.0,
1809
+ "selected_success": 1.0,
1810
+ "selected_success_gain_over_base": 0.0,
1811
+ "selected_utility": 2.0,
1812
+ "selected_utility_gain_over_base": 0.0,
1813
+ "selector_regret": 0.0,
1814
+ "success_selector_gap": 0.0,
1815
+ "success_support_gap": 0.0,
1816
+ "support_gap": 0.0,
1817
+ "tau": 0.29031104002937436
1818
+ },
1819
+ "412102563/train1": {
1820
+ "base_success": 1.0,
1821
+ "base_utility": 2.0,
1822
+ "coverage": 0.0,
1823
+ "execute_generated": 0.0,
1824
+ "fallback_rate": 1.0,
1825
+ "hidden_chart_oracle_success": 1.0,
1826
+ "hidden_chart_oracle_utility": 2.0,
1827
+ "outcome_ptr": 0.0,
1828
+ "predicted_margin": 0.23724048900553812,
1829
+ "proposal_oracle_success": 1.0,
1830
+ "proposal_oracle_utility": 2.0,
1831
+ "selected_candidate_index": 10.0,
1832
+ "selected_success": 1.0,
1833
+ "selected_success_gain_over_base": 0.0,
1834
+ "selected_utility": 2.0,
1835
+ "selected_utility_gain_over_base": 0.0,
1836
+ "selector_regret": 0.0,
1837
+ "success_selector_gap": 0.0,
1838
+ "success_support_gap": 0.0,
1839
+ "support_gap": 0.0,
1840
+ "tau": 0.29031104002937436
1841
+ },
1842
+ "412102563/train2": {
1843
+ "base_success": 1.0,
1844
+ "base_utility": 2.0,
1845
+ "coverage": 0.0,
1846
+ "execute_generated": 0.0,
1847
+ "fallback_rate": 1.0,
1848
+ "hidden_chart_oracle_success": 1.0,
1849
+ "hidden_chart_oracle_utility": 2.0,
1850
+ "outcome_ptr": 0.0,
1851
+ "predicted_margin": 0.21929223189854724,
1852
+ "proposal_oracle_success": 1.0,
1853
+ "proposal_oracle_utility": 2.0,
1854
+ "selected_candidate_index": 4.0,
1855
+ "selected_success": 1.0,
1856
+ "selected_success_gain_over_base": 0.0,
1857
+ "selected_utility": 2.0,
1858
+ "selected_utility_gain_over_base": 0.0,
1859
+ "selector_regret": 0.0,
1860
+ "success_selector_gap": 0.0,
1861
+ "success_support_gap": 0.0,
1862
+ "support_gap": 0.0,
1863
+ "tau": 0.29031104002937436
1864
+ },
1865
+ "440577711/train0": {
1866
+ "base_success": 0.0,
1867
+ "base_utility": 0.036152780055999756,
1868
+ "coverage": 0.0,
1869
+ "execute_generated": 0.0,
1870
+ "fallback_rate": 1.0,
1871
+ "hidden_chart_oracle_success": 1.0,
1872
+ "hidden_chart_oracle_utility": 2.0,
1873
+ "outcome_ptr": 1.0,
1874
+ "predicted_margin": 0.18739562764215434,
1875
+ "proposal_oracle_success": 1.0,
1876
+ "proposal_oracle_utility": 2.0,
1877
+ "selected_candidate_index": 1.0,
1878
+ "selected_success": 0.0,
1879
+ "selected_success_gain_over_base": 0.0,
1880
+ "selected_utility": 0.036152780055999756,
1881
+ "selected_utility_gain_over_base": 0.0,
1882
+ "selector_regret": 1.9638472199440002,
1883
+ "success_selector_gap": 1.0,
1884
+ "success_support_gap": 0.0,
1885
+ "support_gap": 0.0,
1886
+ "tau": 0.29031104002937436
1887
+ },
1888
+ "440577711/train1": {
1889
+ "base_success": 0.0,
1890
+ "base_utility": 0.036152780055999756,
1891
+ "coverage": 0.0,
1892
+ "execute_generated": 0.0,
1893
+ "fallback_rate": 1.0,
1894
+ "hidden_chart_oracle_success": 1.0,
1895
+ "hidden_chart_oracle_utility": 2.0,
1896
+ "outcome_ptr": 1.0,
1897
+ "predicted_margin": 0.18419977674091784,
1898
+ "proposal_oracle_success": 1.0,
1899
+ "proposal_oracle_utility": 2.0,
1900
+ "selected_candidate_index": 1.0,
1901
+ "selected_success": 0.0,
1902
+ "selected_success_gain_over_base": 0.0,
1903
+ "selected_utility": 0.036152780055999756,
1904
+ "selected_utility_gain_over_base": 0.0,
1905
+ "selector_regret": 1.9638472199440002,
1906
+ "success_selector_gap": 1.0,
1907
+ "success_support_gap": 0.0,
1908
+ "support_gap": 0.0,
1909
+ "tau": 0.29031104002937436
1910
+ },
1911
+ "440577711/train2": {
1912
+ "base_success": 0.0,
1913
+ "base_utility": 0.036152780055999756,
1914
+ "coverage": 0.0,
1915
+ "execute_generated": 0.0,
1916
+ "fallback_rate": 1.0,
1917
+ "hidden_chart_oracle_success": 1.0,
1918
+ "hidden_chart_oracle_utility": 2.0,
1919
+ "outcome_ptr": 1.0,
1920
+ "predicted_margin": 0.16442723445279572,
1921
+ "proposal_oracle_success": 1.0,
1922
+ "proposal_oracle_utility": 2.0,
1923
+ "selected_candidate_index": 7.0,
1924
+ "selected_success": 0.0,
1925
+ "selected_success_gain_over_base": 0.0,
1926
+ "selected_utility": 0.036152780055999756,
1927
+ "selected_utility_gain_over_base": 0.0,
1928
+ "selector_regret": 1.9638472199440002,
1929
+ "success_selector_gap": 1.0,
1930
+ "success_support_gap": 0.0,
1931
+ "support_gap": 0.0,
1932
+ "tau": 0.29031104002937436
1933
+ },
1934
+ "441046872/train0": {
1935
+ "base_success": 0.0,
1936
+ "base_utility": 0.6999551653862,
1937
+ "coverage": 1.0,
1938
+ "execute_generated": 1.0,
1939
+ "fallback_rate": 0.0,
1940
+ "hidden_chart_oracle_success": 1.0,
1941
+ "hidden_chart_oracle_utility": 2.0,
1942
+ "outcome_ptr": 1.0,
1943
+ "predicted_margin": 0.33329724085783513,
1944
+ "proposal_oracle_success": 1.0,
1945
+ "proposal_oracle_utility": 2.0,
1946
+ "selected_candidate_index": 10.0,
1947
+ "selected_success": 0.0,
1948
+ "selected_success_gain_over_base": 0.0,
1949
+ "selected_utility": 0.6693241000175476,
1950
+ "selected_utility_gain_over_base": -0.030631065368652344,
1951
+ "selector_regret": 1.3306758999824524,
1952
+ "success_selector_gap": 1.0,
1953
+ "success_support_gap": 0.0,
1954
+ "support_gap": 0.0,
1955
+ "tau": 0.29031104002937436
1956
+ },
1957
+ "441046872/train1": {
1958
+ "base_success": 0.0,
1959
+ "base_utility": 0.6999551653862,
1960
+ "coverage": 1.0,
1961
+ "execute_generated": 1.0,
1962
+ "fallback_rate": 0.0,
1963
+ "hidden_chart_oracle_success": 1.0,
1964
+ "hidden_chart_oracle_utility": 2.0,
1965
+ "outcome_ptr": 1.0,
1966
+ "predicted_margin": 0.3957300144448556,
1967
+ "proposal_oracle_success": 0.0,
1968
+ "proposal_oracle_utility": 0.7233588695526123,
1969
+ "selected_candidate_index": 3.0,
1970
+ "selected_success": 0.0,
1971
+ "selected_success_gain_over_base": 0.0,
1972
+ "selected_utility": 0.663180410861969,
1973
+ "selected_utility_gain_over_base": -0.03677475452423096,
1974
+ "selector_regret": 0.06017845869064331,
1975
+ "success_selector_gap": 0.0,
1976
+ "success_support_gap": 1.0,
1977
+ "support_gap": 1.2766411304473877,
1978
+ "tau": 0.29031104002937436
1979
+ },
1980
+ "441046872/train2": {
1981
+ "base_success": 0.0,
1982
+ "base_utility": 0.6999551653862,
1983
+ "coverage": 1.0,
1984
+ "execute_generated": 1.0,
1985
+ "fallback_rate": 0.0,
1986
+ "hidden_chart_oracle_success": 1.0,
1987
+ "hidden_chart_oracle_utility": 2.0,
1988
+ "outcome_ptr": 1.0,
1989
+ "predicted_margin": 0.33329724085783513,
1990
+ "proposal_oracle_success": 0.0,
1991
+ "proposal_oracle_utility": 0.7191348671913147,
1992
+ "selected_candidate_index": 6.0,
1993
+ "selected_success": 0.0,
1994
+ "selected_success_gain_over_base": 0.0,
1995
+ "selected_utility": 0.6948683261871338,
1996
+ "selected_utility_gain_over_base": -0.005086839199066162,
1997
+ "selector_regret": 0.024266541004180908,
1998
+ "success_selector_gap": 0.0,
1999
+ "success_support_gap": 1.0,
2000
+ "support_gap": 1.2808651328086853,
2001
+ "tau": 0.29031104002937436
2002
+ },
2003
+ "509430392/train0": {
2004
+ "base_success": 0.0,
2005
+ "base_utility": 2.2351741790771484e-07,
2006
+ "coverage": 1.0,
2007
+ "execute_generated": 1.0,
2008
+ "fallback_rate": 0.0,
2009
+ "hidden_chart_oracle_success": 1.0,
2010
+ "hidden_chart_oracle_utility": 2.0,
2011
+ "outcome_ptr": 1.0,
2012
+ "predicted_margin": 0.5951794014623365,
2013
+ "proposal_oracle_success": 1.0,
2014
+ "proposal_oracle_utility": 2.0,
2015
+ "selected_candidate_index": 14.0,
2016
+ "selected_success": 0.0,
2017
+ "selected_success_gain_over_base": 0.0,
2018
+ "selected_utility": 0.12097799777984619,
2019
+ "selected_utility_gain_over_base": 0.12097777426242828,
2020
+ "selector_regret": 1.8790220022201538,
2021
+ "success_selector_gap": 1.0,
2022
+ "success_support_gap": 0.0,
2023
+ "support_gap": 0.0,
2024
+ "tau": 0.29031104002937436
2025
+ },
2026
+ "509430392/train1": {
2027
+ "base_success": 0.0,
2028
+ "base_utility": 2.2351741790771484e-07,
2029
+ "coverage": 1.0,
2030
+ "execute_generated": 1.0,
2031
+ "fallback_rate": 0.0,
2032
+ "hidden_chart_oracle_success": 1.0,
2033
+ "hidden_chart_oracle_utility": 2.0,
2034
+ "outcome_ptr": 1.0,
2035
+ "predicted_margin": 0.5908952862576345,
2036
+ "proposal_oracle_success": 1.0,
2037
+ "proposal_oracle_utility": 2.0,
2038
+ "selected_candidate_index": 3.0,
2039
+ "selected_success": 1.0,
2040
+ "selected_success_gain_over_base": 1.0,
2041
+ "selected_utility": 2.0,
2042
+ "selected_utility_gain_over_base": 1.999999776482582,
2043
+ "selector_regret": 0.0,
2044
+ "success_selector_gap": 0.0,
2045
+ "success_support_gap": 0.0,
2046
+ "support_gap": 0.0,
2047
+ "tau": 0.29031104002937436
2048
+ },
2049
+ "509430392/train2": {
2050
+ "base_success": 0.0,
2051
+ "base_utility": 2.2351741790771484e-07,
2052
+ "coverage": 1.0,
2053
+ "execute_generated": 1.0,
2054
+ "fallback_rate": 0.0,
2055
+ "hidden_chart_oracle_success": 1.0,
2056
+ "hidden_chart_oracle_utility": 2.0,
2057
+ "outcome_ptr": 1.0,
2058
+ "predicted_margin": 0.6194446763134166,
2059
+ "proposal_oracle_success": 0.0,
2060
+ "proposal_oracle_utility": 5.424022674560547e-06,
2061
+ "selected_candidate_index": 10.0,
2062
+ "selected_success": 0.0,
2063
+ "selected_success_gain_over_base": 0.0,
2064
+ "selected_utility": 5.424022674560547e-06,
2065
+ "selected_utility_gain_over_base": 5.200505256652832e-06,
2066
+ "selector_regret": 0.0,
2067
+ "success_selector_gap": 0.0,
2068
+ "success_support_gap": 1.0,
2069
+ "support_gap": 1.9999945759773254,
2070
+ "tau": 0.29031104002937436
2071
+ },
2072
+ "585904130/train0": {
2073
+ "base_success": 0.0,
2074
+ "base_utility": 0.18171554803848267,
2075
+ "coverage": 1.0,
2076
+ "execute_generated": 1.0,
2077
+ "fallback_rate": 0.0,
2078
+ "hidden_chart_oracle_success": 1.0,
2079
+ "hidden_chart_oracle_utility": 2.0,
2080
+ "outcome_ptr": 1.0,
2081
+ "predicted_margin": 0.3083190727440344,
2082
+ "proposal_oracle_success": 1.0,
2083
+ "proposal_oracle_utility": 2.0,
2084
+ "selected_candidate_index": 1.0,
2085
+ "selected_success": 0.0,
2086
+ "selected_success_gain_over_base": 0.0,
2087
+ "selected_utility": 0.05923032760620117,
2088
+ "selected_utility_gain_over_base": -0.1224852204322815,
2089
+ "selector_regret": 1.9407696723937988,
2090
+ "success_selector_gap": 1.0,
2091
+ "success_support_gap": 0.0,
2092
+ "support_gap": 0.0,
2093
+ "tau": 0.29031104002937436
2094
+ },
2095
+ "585904130/train1": {
2096
+ "base_success": 0.0,
2097
+ "base_utility": 0.18171554803848267,
2098
+ "coverage": 1.0,
2099
+ "execute_generated": 1.0,
2100
+ "fallback_rate": 0.0,
2101
+ "hidden_chart_oracle_success": 1.0,
2102
+ "hidden_chart_oracle_utility": 2.0,
2103
+ "outcome_ptr": 1.0,
2104
+ "predicted_margin": 0.3177387291157021,
2105
+ "proposal_oracle_success": 1.0,
2106
+ "proposal_oracle_utility": 2.0,
2107
+ "selected_candidate_index": 0.0,
2108
+ "selected_success": 1.0,
2109
+ "selected_success_gain_over_base": 1.0,
2110
+ "selected_utility": 2.0,
2111
+ "selected_utility_gain_over_base": 1.8182844519615173,
2112
+ "selector_regret": 0.0,
2113
+ "success_selector_gap": 0.0,
2114
+ "success_support_gap": 0.0,
2115
+ "support_gap": 0.0,
2116
+ "tau": 0.29031104002937436
2117
+ },
2118
+ "585904130/train2": {
2119
+ "base_success": 0.0,
2120
+ "base_utility": 0.18171554803848267,
2121
+ "coverage": 0.0,
2122
+ "execute_generated": 0.0,
2123
+ "fallback_rate": 1.0,
2124
+ "hidden_chart_oracle_success": 1.0,
2125
+ "hidden_chart_oracle_utility": 2.0,
2126
+ "outcome_ptr": 1.0,
2127
+ "predicted_margin": 0.2046539172174809,
2128
+ "proposal_oracle_success": 1.0,
2129
+ "proposal_oracle_utility": 2.0,
2130
+ "selected_candidate_index": 0.0,
2131
+ "selected_success": 0.0,
2132
+ "selected_success_gain_over_base": 0.0,
2133
+ "selected_utility": 0.18171554803848267,
2134
+ "selected_utility_gain_over_base": 0.0,
2135
+ "selector_regret": 1.8182844519615173,
2136
+ "success_selector_gap": 1.0,
2137
+ "success_support_gap": 0.0,
2138
+ "support_gap": 0.0,
2139
+ "tau": 0.29031104002937436
2140
+ },
2141
+ "608016290/train0": {
2142
+ "base_success": 0.0,
2143
+ "base_utility": 0.725701630115509,
2144
+ "coverage": 0.0,
2145
+ "execute_generated": 0.0,
2146
+ "fallback_rate": 1.0,
2147
+ "hidden_chart_oracle_success": 1.0,
2148
+ "hidden_chart_oracle_utility": 2.0,
2149
+ "outcome_ptr": 1.0,
2150
+ "predicted_margin": 0.1565404769152266,
2151
+ "proposal_oracle_success": 1.0,
2152
+ "proposal_oracle_utility": 2.0,
2153
+ "selected_candidate_index": 15.0,
2154
+ "selected_success": 0.0,
2155
+ "selected_success_gain_over_base": 0.0,
2156
+ "selected_utility": 0.725701630115509,
2157
+ "selected_utility_gain_over_base": 0.0,
2158
+ "selector_regret": 1.274298369884491,
2159
+ "success_selector_gap": 1.0,
2160
+ "success_support_gap": 0.0,
2161
+ "support_gap": 0.0,
2162
+ "tau": 0.29031104002937436
2163
+ },
2164
+ "608016290/train1": {
2165
+ "base_success": 0.0,
2166
+ "base_utility": 0.725701630115509,
2167
+ "coverage": 0.0,
2168
+ "execute_generated": 0.0,
2169
+ "fallback_rate": 1.0,
2170
+ "hidden_chart_oracle_success": 1.0,
2171
+ "hidden_chart_oracle_utility": 2.0,
2172
+ "outcome_ptr": 1.0,
2173
+ "predicted_margin": 0.1646347051814787,
2174
+ "proposal_oracle_success": 1.0,
2175
+ "proposal_oracle_utility": 2.0,
2176
+ "selected_candidate_index": 9.0,
2177
+ "selected_success": 0.0,
2178
+ "selected_success_gain_over_base": 0.0,
2179
+ "selected_utility": 0.725701630115509,
2180
+ "selected_utility_gain_over_base": 0.0,
2181
+ "selector_regret": 1.274298369884491,
2182
+ "success_selector_gap": 1.0,
2183
+ "success_support_gap": 0.0,
2184
+ "support_gap": 0.0,
2185
+ "tau": 0.29031104002937436
2186
+ },
2187
+ "608016290/train2": {
2188
+ "base_success": 0.0,
2189
+ "base_utility": 0.725701630115509,
2190
+ "coverage": 0.0,
2191
+ "execute_generated": 0.0,
2192
+ "fallback_rate": 1.0,
2193
+ "hidden_chart_oracle_success": 1.0,
2194
+ "hidden_chart_oracle_utility": 2.0,
2195
+ "outcome_ptr": 0.0,
2196
+ "predicted_margin": 0.15758444275633177,
2197
+ "proposal_oracle_success": 0.0,
2198
+ "proposal_oracle_utility": 0.7117276191711426,
2199
+ "selected_candidate_index": 9.0,
2200
+ "selected_success": 0.0,
2201
+ "selected_success_gain_over_base": 0.0,
2202
+ "selected_utility": 0.725701630115509,
2203
+ "selected_utility_gain_over_base": 0.0,
2204
+ "selector_regret": 0.0,
2205
+ "success_selector_gap": 0.0,
2206
+ "success_support_gap": 1.0,
2207
+ "support_gap": 1.2882723808288574,
2208
+ "tau": 0.29031104002937436
2209
+ },
2210
+ "626486813/train0": {
2211
+ "base_success": 0.0,
2212
+ "base_utility": 0.00509834272088483,
2213
+ "coverage": 0.5,
2214
+ "execute_generated": 0.5,
2215
+ "fallback_rate": 0.5,
2216
+ "hidden_chart_oracle_success": 0.0,
2217
+ "hidden_chart_oracle_utility": 0.16558238118886948,
2218
+ "outcome_ptr": 1.0,
2219
+ "predicted_margin": 0.5564234962339261,
2220
+ "proposal_oracle_success": 0.0,
2221
+ "proposal_oracle_utility": 0.04469547560438514,
2222
+ "selected_candidate_index": 1.5,
2223
+ "selected_success": 0.0,
2224
+ "selected_success_gain_over_base": 0.0,
2225
+ "selected_utility": 0.005386819248087704,
2226
+ "selected_utility_gain_over_base": 0.0002884765272028744,
2227
+ "selector_regret": 0.03930865635629743,
2228
+ "success_selector_gap": 0.0,
2229
+ "success_support_gap": 0.0,
2230
+ "support_gap": 0.12161866808310151,
2231
+ "tau": 0.29031104002937436
2232
+ },
2233
+ "626486813/train1": {
2234
+ "base_success": 0.0,
2235
+ "base_utility": 0.00509834272088483,
2236
+ "coverage": 1.0,
2237
+ "execute_generated": 1.0,
2238
+ "fallback_rate": 0.0,
2239
+ "hidden_chart_oracle_success": 0.0,
2240
+ "hidden_chart_oracle_utility": 0.16558238118886948,
2241
+ "outcome_ptr": 1.0,
2242
+ "predicted_margin": 0.5565345976831662,
2243
+ "proposal_oracle_success": 0.0,
2244
+ "proposal_oracle_utility": 0.035827874671667814,
2245
+ "selected_candidate_index": 0.5,
2246
+ "selected_success": 0.0,
2247
+ "selected_success_gain_over_base": 0.0,
2248
+ "selected_utility": 0.02161153219640255,
2249
+ "selected_utility_gain_over_base": 0.01651318947551772,
2250
+ "selector_regret": 0.014216342475265265,
2251
+ "success_selector_gap": 0.0,
2252
+ "success_support_gap": 0.0,
2253
+ "support_gap": 0.12975450651720166,
2254
+ "tau": 0.29031104002937436
2255
+ },
2256
+ "626486813/train2": {
2257
+ "base_success": 0.0,
2258
+ "base_utility": 0.00509834272088483,
2259
+ "coverage": 0.5,
2260
+ "execute_generated": 0.5,
2261
+ "fallback_rate": 0.5,
2262
+ "hidden_chart_oracle_success": 0.0,
2263
+ "hidden_chart_oracle_utility": 0.16558238118886948,
2264
+ "outcome_ptr": 1.0,
2265
+ "predicted_margin": 0.47780405773180123,
2266
+ "proposal_oracle_success": 0.0,
2267
+ "proposal_oracle_utility": 0.04489925503730774,
2268
+ "selected_candidate_index": 1.0,
2269
+ "selected_success": 0.0,
2270
+ "selected_success_gain_over_base": 0.0,
2271
+ "selected_utility": 0.015375842805951834,
2272
+ "selected_utility_gain_over_base": 0.010277500085067004,
2273
+ "selector_regret": 0.029523412231355906,
2274
+ "success_selector_gap": 0.0,
2275
+ "success_support_gap": 0.0,
2276
+ "support_gap": 0.12068312615156174,
2277
+ "tau": 0.29031104002937436
2278
+ },
2279
+ "636680137/train0": {
2280
+ "base_success": 0.0,
2281
+ "base_utility": 0.08313504606485367,
2282
+ "coverage": 1.0,
2283
+ "execute_generated": 1.0,
2284
+ "fallback_rate": 0.0,
2285
+ "hidden_chart_oracle_success": 0.0,
2286
+ "hidden_chart_oracle_utility": 0.17056432366371155,
2287
+ "outcome_ptr": 1.0,
2288
+ "predicted_margin": 0.40225098671598103,
2289
+ "proposal_oracle_success": 0.0,
2290
+ "proposal_oracle_utility": 0.08826130628585815,
2291
+ "selected_candidate_index": 0.0,
2292
+ "selected_success": 0.0,
2293
+ "selected_success_gain_over_base": 0.0,
2294
+ "selected_utility": 0.056092143058776855,
2295
+ "selected_utility_gain_over_base": -0.027042903006076813,
2296
+ "selector_regret": 0.0321691632270813,
2297
+ "success_selector_gap": 0.0,
2298
+ "success_support_gap": 0.0,
2299
+ "support_gap": 0.0823030173778534,
2300
+ "tau": 0.29031104002937436
2301
+ },
2302
+ "636680137/train1": {
2303
+ "base_success": 0.0,
2304
+ "base_utility": 0.08313504606485367,
2305
+ "coverage": 0.0,
2306
+ "execute_generated": 0.0,
2307
+ "fallback_rate": 1.0,
2308
+ "hidden_chart_oracle_success": 0.0,
2309
+ "hidden_chart_oracle_utility": 0.17056432366371155,
2310
+ "outcome_ptr": 0.0,
2311
+ "predicted_margin": 0.18624964231819074,
2312
+ "proposal_oracle_success": 0.0,
2313
+ "proposal_oracle_utility": 0.06741118431091309,
2314
+ "selected_candidate_index": 2.0,
2315
+ "selected_success": 0.0,
2316
+ "selected_success_gain_over_base": 0.0,
2317
+ "selected_utility": 0.08313504606485367,
2318
+ "selected_utility_gain_over_base": 0.0,
2319
+ "selector_regret": 0.0,
2320
+ "success_selector_gap": 0.0,
2321
+ "success_support_gap": 0.0,
2322
+ "support_gap": 0.10315313935279846,
2323
+ "tau": 0.29031104002937436
2324
+ },
2325
+ "636680137/train2": {
2326
+ "base_success": 0.0,
2327
+ "base_utility": 0.08313504606485367,
2328
+ "coverage": 0.0,
2329
+ "execute_generated": 0.0,
2330
+ "fallback_rate": 1.0,
2331
+ "hidden_chart_oracle_success": 0.0,
2332
+ "hidden_chart_oracle_utility": 0.17056432366371155,
2333
+ "outcome_ptr": 0.0,
2334
+ "predicted_margin": 0.2257704957099745,
2335
+ "proposal_oracle_success": 0.0,
2336
+ "proposal_oracle_utility": 0.0756516233086586,
2337
+ "selected_candidate_index": 0.0,
2338
+ "selected_success": 0.0,
2339
+ "selected_success_gain_over_base": 0.0,
2340
+ "selected_utility": 0.08313504606485367,
2341
+ "selected_utility_gain_over_base": 0.0,
2342
+ "selector_regret": 0.0,
2343
+ "success_selector_gap": 0.0,
2344
+ "success_support_gap": 0.0,
2345
+ "support_gap": 0.09491270035505295,
2346
+ "tau": 0.29031104002937436
2347
+ },
2348
+ "655884605/train0": {
2349
+ "base_success": 0.0,
2350
+ "base_utility": 0.16300667822360992,
2351
+ "coverage": 0.0,
2352
+ "execute_generated": 0.0,
2353
+ "fallback_rate": 1.0,
2354
+ "hidden_chart_oracle_success": 1.0,
2355
+ "hidden_chart_oracle_utility": 2.0,
2356
+ "outcome_ptr": 0.0,
2357
+ "predicted_margin": 0.26463671358858426,
2358
+ "proposal_oracle_success": 0.0,
2359
+ "proposal_oracle_utility": 0.13732868432998657,
2360
+ "selected_candidate_index": 5.0,
2361
+ "selected_success": 0.0,
2362
+ "selected_success_gain_over_base": 0.0,
2363
+ "selected_utility": 0.16300667822360992,
2364
+ "selected_utility_gain_over_base": 0.0,
2365
+ "selector_regret": 0.0,
2366
+ "success_selector_gap": 0.0,
2367
+ "success_support_gap": 1.0,
2368
+ "support_gap": 1.8626713156700134,
2369
+ "tau": 0.29031104002937436
2370
+ },
2371
+ "655884605/train1": {
2372
+ "base_success": 0.0,
2373
+ "base_utility": 0.16300667822360992,
2374
+ "coverage": 0.0,
2375
+ "execute_generated": 0.0,
2376
+ "fallback_rate": 1.0,
2377
+ "hidden_chart_oracle_success": 1.0,
2378
+ "hidden_chart_oracle_utility": 2.0,
2379
+ "outcome_ptr": 1.0,
2380
+ "predicted_margin": 0.24600950331932453,
2381
+ "proposal_oracle_success": 0.0,
2382
+ "proposal_oracle_utility": 0.8128758072853088,
2383
+ "selected_candidate_index": 2.0,
2384
+ "selected_success": 0.0,
2385
+ "selected_success_gain_over_base": 0.0,
2386
+ "selected_utility": 0.16300667822360992,
2387
+ "selected_utility_gain_over_base": 0.0,
2388
+ "selector_regret": 0.6498691290616989,
2389
+ "success_selector_gap": 0.0,
2390
+ "success_support_gap": 1.0,
2391
+ "support_gap": 1.1871241927146912,
2392
+ "tau": 0.29031104002937436
2393
+ },
2394
+ "655884605/train2": {
2395
+ "base_success": 0.0,
2396
+ "base_utility": 0.16300667822360992,
2397
+ "coverage": 0.0,
2398
+ "execute_generated": 0.0,
2399
+ "fallback_rate": 1.0,
2400
+ "hidden_chart_oracle_success": 1.0,
2401
+ "hidden_chart_oracle_utility": 2.0,
2402
+ "outcome_ptr": 1.0,
2403
+ "predicted_margin": 0.181775355536573,
2404
+ "proposal_oracle_success": 0.0,
2405
+ "proposal_oracle_utility": 0.21447330713272095,
2406
+ "selected_candidate_index": 3.0,
2407
+ "selected_success": 0.0,
2408
+ "selected_success_gain_over_base": 0.0,
2409
+ "selected_utility": 0.16300667822360992,
2410
+ "selected_utility_gain_over_base": 0.0,
2411
+ "selector_regret": 0.05146662890911102,
2412
+ "success_selector_gap": 0.0,
2413
+ "success_support_gap": 1.0,
2414
+ "support_gap": 1.785526692867279,
2415
+ "tau": 0.29031104002937436
2416
+ },
2417
+ "672328262/train0": {
2418
+ "base_success": 1.0,
2419
+ "base_utility": 2.0,
2420
+ "coverage": 0.0,
2421
+ "execute_generated": 0.0,
2422
+ "fallback_rate": 1.0,
2423
+ "hidden_chart_oracle_success": 1.0,
2424
+ "hidden_chart_oracle_utility": 2.0,
2425
+ "outcome_ptr": 0.0,
2426
+ "predicted_margin": 0.04820260893743538,
2427
+ "proposal_oracle_success": 1.0,
2428
+ "proposal_oracle_utility": 2.0,
2429
+ "selected_candidate_index": 2.0,
2430
+ "selected_success": 1.0,
2431
+ "selected_success_gain_over_base": 0.0,
2432
+ "selected_utility": 2.0,
2433
+ "selected_utility_gain_over_base": 0.0,
2434
+ "selector_regret": 0.0,
2435
+ "success_selector_gap": 0.0,
2436
+ "success_support_gap": 0.0,
2437
+ "support_gap": 0.0,
2438
+ "tau": 0.29031104002937436
2439
+ },
2440
+ "672328262/train1": {
2441
+ "base_success": 1.0,
2442
+ "base_utility": 2.0,
2443
+ "coverage": 0.0,
2444
+ "execute_generated": 0.0,
2445
+ "fallback_rate": 1.0,
2446
+ "hidden_chart_oracle_success": 1.0,
2447
+ "hidden_chart_oracle_utility": 2.0,
2448
+ "outcome_ptr": 0.0,
2449
+ "predicted_margin": 0.05319974288197324,
2450
+ "proposal_oracle_success": 1.0,
2451
+ "proposal_oracle_utility": 2.0,
2452
+ "selected_candidate_index": 9.0,
2453
+ "selected_success": 1.0,
2454
+ "selected_success_gain_over_base": 0.0,
2455
+ "selected_utility": 2.0,
2456
+ "selected_utility_gain_over_base": 0.0,
2457
+ "selector_regret": 0.0,
2458
+ "success_selector_gap": 0.0,
2459
+ "success_support_gap": 0.0,
2460
+ "support_gap": 0.0,
2461
+ "tau": 0.29031104002937436
2462
+ },
2463
+ "672328262/train2": {
2464
+ "base_success": 1.0,
2465
+ "base_utility": 2.0,
2466
+ "coverage": 0.0,
2467
+ "execute_generated": 0.0,
2468
+ "fallback_rate": 1.0,
2469
+ "hidden_chart_oracle_success": 1.0,
2470
+ "hidden_chart_oracle_utility": 2.0,
2471
+ "outcome_ptr": 0.0,
2472
+ "predicted_margin": 0.047409745226937555,
2473
+ "proposal_oracle_success": 1.0,
2474
+ "proposal_oracle_utility": 2.0,
2475
+ "selected_candidate_index": 7.0,
2476
+ "selected_success": 1.0,
2477
+ "selected_success_gain_over_base": 0.0,
2478
+ "selected_utility": 2.0,
2479
+ "selected_utility_gain_over_base": 0.0,
2480
+ "selector_regret": 0.0,
2481
+ "success_selector_gap": 0.0,
2482
+ "success_support_gap": 0.0,
2483
+ "support_gap": 0.0,
2484
+ "tau": 0.29031104002937436
2485
+ },
2486
+ "701667486/train0": {
2487
+ "base_success": 1.0,
2488
+ "base_utility": 2.0,
2489
+ "coverage": 1.0,
2490
+ "execute_generated": 1.0,
2491
+ "fallback_rate": 0.0,
2492
+ "hidden_chart_oracle_success": 1.0,
2493
+ "hidden_chart_oracle_utility": 2.0,
2494
+ "outcome_ptr": 0.0,
2495
+ "predicted_margin": 0.5039098400899894,
2496
+ "proposal_oracle_success": 1.0,
2497
+ "proposal_oracle_utility": 2.0,
2498
+ "selected_candidate_index": 7.0,
2499
+ "selected_success": 1.0,
2500
+ "selected_success_gain_over_base": 0.0,
2501
+ "selected_utility": 2.0,
2502
+ "selected_utility_gain_over_base": 0.0,
2503
+ "selector_regret": 0.0,
2504
+ "success_selector_gap": 0.0,
2505
+ "success_support_gap": 0.0,
2506
+ "support_gap": 0.0,
2507
+ "tau": 0.29031104002937436
2508
+ },
2509
+ "701667486/train1": {
2510
+ "base_success": 1.0,
2511
+ "base_utility": 2.0,
2512
+ "coverage": 1.0,
2513
+ "execute_generated": 1.0,
2514
+ "fallback_rate": 0.0,
2515
+ "hidden_chart_oracle_success": 1.0,
2516
+ "hidden_chart_oracle_utility": 2.0,
2517
+ "outcome_ptr": 0.0,
2518
+ "predicted_margin": 0.4886226110342108,
2519
+ "proposal_oracle_success": 1.0,
2520
+ "proposal_oracle_utility": 2.0,
2521
+ "selected_candidate_index": 7.0,
2522
+ "selected_success": 0.0,
2523
+ "selected_success_gain_over_base": -1.0,
2524
+ "selected_utility": 0.8246148228645325,
2525
+ "selected_utility_gain_over_base": -1.1753851771354675,
2526
+ "selector_regret": 1.1753851771354675,
2527
+ "success_selector_gap": 1.0,
2528
+ "success_support_gap": 0.0,
2529
+ "support_gap": 0.0,
2530
+ "tau": 0.29031104002937436
2531
+ },
2532
+ "701667486/train2": {
2533
+ "base_success": 1.0,
2534
+ "base_utility": 2.0,
2535
+ "coverage": 1.0,
2536
+ "execute_generated": 1.0,
2537
+ "fallback_rate": 0.0,
2538
+ "hidden_chart_oracle_success": 1.0,
2539
+ "hidden_chart_oracle_utility": 2.0,
2540
+ "outcome_ptr": 0.0,
2541
+ "predicted_margin": 0.4915250634255605,
2542
+ "proposal_oracle_success": 1.0,
2543
+ "proposal_oracle_utility": 2.0,
2544
+ "selected_candidate_index": 5.0,
2545
+ "selected_success": 1.0,
2546
+ "selected_success_gain_over_base": 0.0,
2547
+ "selected_utility": 2.0,
2548
+ "selected_utility_gain_over_base": 0.0,
2549
+ "selector_regret": 0.0,
2550
+ "success_selector_gap": 0.0,
2551
+ "success_support_gap": 0.0,
2552
+ "support_gap": 0.0,
2553
+ "tau": 0.29031104002937436
2554
+ },
2555
+ "747260437/train0": {
2556
+ "base_success": 1.0,
2557
+ "base_utility": 2.0,
2558
+ "coverage": 0.0,
2559
+ "execute_generated": 0.0,
2560
+ "fallback_rate": 1.0,
2561
+ "hidden_chart_oracle_success": 0.0,
2562
+ "hidden_chart_oracle_utility": 0.2786632478237152,
2563
+ "outcome_ptr": 0.0,
2564
+ "predicted_margin": 0.1675764505623386,
2565
+ "proposal_oracle_success": 0.0,
2566
+ "proposal_oracle_utility": 0.10839563608169556,
2567
+ "selected_candidate_index": 0.0,
2568
+ "selected_success": 1.0,
2569
+ "selected_success_gain_over_base": 0.0,
2570
+ "selected_utility": 2.0,
2571
+ "selected_utility_gain_over_base": 0.0,
2572
+ "selector_regret": 0.0,
2573
+ "success_selector_gap": 0.0,
2574
+ "success_support_gap": 0.0,
2575
+ "support_gap": 0.17026761174201965,
2576
+ "tau": 0.29031104002937436
2577
+ },
2578
+ "747260437/train1": {
2579
+ "base_success": 1.0,
2580
+ "base_utility": 2.0,
2581
+ "coverage": 0.0,
2582
+ "execute_generated": 0.0,
2583
+ "fallback_rate": 1.0,
2584
+ "hidden_chart_oracle_success": 0.0,
2585
+ "hidden_chart_oracle_utility": 0.2786632478237152,
2586
+ "outcome_ptr": 0.0,
2587
+ "predicted_margin": 0.10738897055995479,
2588
+ "proposal_oracle_success": 0.0,
2589
+ "proposal_oracle_utility": 0.06328952312469482,
2590
+ "selected_candidate_index": 8.0,
2591
+ "selected_success": 1.0,
2592
+ "selected_success_gain_over_base": 0.0,
2593
+ "selected_utility": 2.0,
2594
+ "selected_utility_gain_over_base": 0.0,
2595
+ "selector_regret": 0.0,
2596
+ "success_selector_gap": 0.0,
2597
+ "success_support_gap": 0.0,
2598
+ "support_gap": 0.21537372469902039,
2599
+ "tau": 0.29031104002937436
2600
+ },
2601
+ "747260437/train2": {
2602
+ "base_success": 1.0,
2603
+ "base_utility": 2.0,
2604
+ "coverage": 0.0,
2605
+ "execute_generated": 0.0,
2606
+ "fallback_rate": 1.0,
2607
+ "hidden_chart_oracle_success": 0.0,
2608
+ "hidden_chart_oracle_utility": 0.2786632478237152,
2609
+ "outcome_ptr": 0.0,
2610
+ "predicted_margin": 0.21986396235622255,
2611
+ "proposal_oracle_success": 0.0,
2612
+ "proposal_oracle_utility": 0.12510187923908234,
2613
+ "selected_candidate_index": 1.0,
2614
+ "selected_success": 1.0,
2615
+ "selected_success_gain_over_base": 0.0,
2616
+ "selected_utility": 2.0,
2617
+ "selected_utility_gain_over_base": 0.0,
2618
+ "selector_regret": 0.0,
2619
+ "success_selector_gap": 0.0,
2620
+ "success_support_gap": 0.0,
2621
+ "support_gap": 0.15356136858463287,
2622
+ "tau": 0.29031104002937436
2623
+ },
2624
+ "750448296/train0": {
2625
+ "base_success": 0.0,
2626
+ "base_utility": 0.11759314934412639,
2627
+ "coverage": 0.3333333333333333,
2628
+ "execute_generated": 0.3333333333333333,
2629
+ "fallback_rate": 0.6666666666666666,
2630
+ "hidden_chart_oracle_success": 0.6666666666666666,
2631
+ "hidden_chart_oracle_utility": 1.4079964955647786,
2632
+ "outcome_ptr": 0.6666666666666666,
2633
+ "predicted_margin": 0.37880623603795066,
2634
+ "proposal_oracle_success": 0.3333333333333333,
2635
+ "proposal_oracle_utility": 0.7526603279014429,
2636
+ "selected_candidate_index": 3.3333333333333335,
2637
+ "selected_success": 0.0,
2638
+ "selected_success_gain_over_base": 0.0,
2639
+ "selected_utility": 0.1003921168545882,
2640
+ "selected_utility_gain_over_base": -0.017201032489538193,
2641
+ "selector_regret": 0.6522682110468546,
2642
+ "success_selector_gap": 0.3333333333333333,
2643
+ "success_support_gap": 0.3333333333333333,
2644
+ "support_gap": 0.6553361676633358,
2645
+ "tau": 0.29031104002937436
2646
+ },
2647
+ "750448296/train1": {
2648
+ "base_success": 0.0,
2649
+ "base_utility": 0.11759314934412639,
2650
+ "coverage": 0.3333333333333333,
2651
+ "execute_generated": 0.3333333333333333,
2652
+ "fallback_rate": 0.6666666666666666,
2653
+ "hidden_chart_oracle_success": 0.6666666666666666,
2654
+ "hidden_chart_oracle_utility": 1.4079964955647786,
2655
+ "outcome_ptr": 1.0,
2656
+ "predicted_margin": 0.34807137553236295,
2657
+ "proposal_oracle_success": 0.6666666666666666,
2658
+ "proposal_oracle_utility": 1.3705032741030057,
2659
+ "selected_candidate_index": 8.333333333333334,
2660
+ "selected_success": 0.0,
2661
+ "selected_success_gain_over_base": 0.0,
2662
+ "selected_utility": 0.11465192586183548,
2663
+ "selected_utility_gain_over_base": -0.0029412234822909036,
2664
+ "selector_regret": 1.2558513482411702,
2665
+ "success_selector_gap": 0.6666666666666666,
2666
+ "success_support_gap": 0.0,
2667
+ "support_gap": 0.03749322146177292,
2668
+ "tau": 0.29031104002937436
2669
+ },
2670
+ "750448296/train2": {
2671
+ "base_success": 0.0,
2672
+ "base_utility": 0.11759314934412639,
2673
+ "coverage": 0.3333333333333333,
2674
+ "execute_generated": 0.3333333333333333,
2675
+ "fallback_rate": 0.6666666666666666,
2676
+ "hidden_chart_oracle_success": 0.6666666666666666,
2677
+ "hidden_chart_oracle_utility": 1.4079964955647786,
2678
+ "outcome_ptr": 0.6666666666666666,
2679
+ "predicted_margin": 0.32713744779683995,
2680
+ "proposal_oracle_success": 0.6666666666666666,
2681
+ "proposal_oracle_utility": 1.3504486680030823,
2682
+ "selected_candidate_index": 5.333333333333333,
2683
+ "selected_success": 0.0,
2684
+ "selected_success_gain_over_base": 0.0,
2685
+ "selected_utility": 0.1037093202273051,
2686
+ "selected_utility_gain_over_base": -0.013883829116821289,
2687
+ "selector_regret": 1.246739347775777,
2688
+ "success_selector_gap": 0.6666666666666666,
2689
+ "success_support_gap": 0.0,
2690
+ "support_gap": 0.05754782756169637,
2691
+ "tau": 0.29031104002937436
2692
+ },
2693
+ "767747424/train0": {
2694
+ "base_success": 0.25,
2695
+ "base_utility": 0.5731993420049548,
2696
+ "coverage": 0.5,
2697
+ "execute_generated": 0.5,
2698
+ "fallback_rate": 0.5,
2699
+ "hidden_chart_oracle_success": 0.25,
2700
+ "hidden_chart_oracle_utility": 0.6737305596470833,
2701
+ "outcome_ptr": 0.5,
2702
+ "predicted_margin": 0.2970760855424009,
2703
+ "proposal_oracle_success": 0.5,
2704
+ "proposal_oracle_utility": 1.0424291878007352,
2705
+ "selected_candidate_index": 9.5,
2706
+ "selected_success": 0.25,
2707
+ "selected_success_gain_over_base": 0.0,
2708
+ "selected_utility": 0.5294794533401728,
2709
+ "selected_utility_gain_over_base": -0.04371988866478205,
2710
+ "selector_regret": 0.5129497344605625,
2711
+ "success_selector_gap": 0.25,
2712
+ "success_support_gap": 0.0,
2713
+ "support_gap": 0.06562284985557199,
2714
+ "tau": 0.29031104002937436
2715
+ },
2716
+ "767747424/train1": {
2717
+ "base_success": 0.25,
2718
+ "base_utility": 0.5731993420049548,
2719
+ "coverage": 0.25,
2720
+ "execute_generated": 0.25,
2721
+ "fallback_rate": 0.75,
2722
+ "hidden_chart_oracle_success": 0.25,
2723
+ "hidden_chart_oracle_utility": 0.6737305596470833,
2724
+ "outcome_ptr": 0.5,
2725
+ "predicted_margin": 0.2643064870157273,
2726
+ "proposal_oracle_success": 0.5,
2727
+ "proposal_oracle_utility": 1.0629530912265182,
2728
+ "selected_candidate_index": 8.0,
2729
+ "selected_success": 0.25,
2730
+ "selected_success_gain_over_base": 0.0,
2731
+ "selected_utility": 0.562042286619544,
2732
+ "selected_utility_gain_over_base": -0.011157055385410786,
2733
+ "selector_regret": 0.5009108046069741,
2734
+ "success_selector_gap": 0.25,
2735
+ "success_support_gap": 0.0,
2736
+ "support_gap": 0.045098946429789066,
2737
+ "tau": 0.29031104002937436
2738
+ },
2739
+ "767747424/train2": {
2740
+ "base_success": 0.25,
2741
+ "base_utility": 0.5731993420049548,
2742
+ "coverage": 0.25,
2743
+ "execute_generated": 0.25,
2744
+ "fallback_rate": 0.75,
2745
+ "hidden_chart_oracle_success": 0.25,
2746
+ "hidden_chart_oracle_utility": 0.6737305596470833,
2747
+ "outcome_ptr": 0.5,
2748
+ "predicted_margin": 0.25424848075521866,
2749
+ "proposal_oracle_success": 0.5,
2750
+ "proposal_oracle_utility": 1.0671670949086547,
2751
+ "selected_candidate_index": 5.25,
2752
+ "selected_success": 0.25,
2753
+ "selected_success_gain_over_base": 0.0,
2754
+ "selected_utility": 0.572120369412005,
2755
+ "selected_utility_gain_over_base": -0.0010789725929498672,
2756
+ "selector_regret": 0.49504672549664974,
2757
+ "success_selector_gap": 0.25,
2758
+ "success_support_gap": 0.0,
2759
+ "support_gap": 0.04088494274765253,
2760
+ "tau": 0.29031104002937436
2761
+ }
2762
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/metrics_by_task.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "LiftPegUpright-v1": {
3
+ "base_success": 0.2,
4
+ "base_utility": 0.9498705387115478,
5
+ "coverage": 0.4666666666666667,
6
+ "execute_generated": 0.4666666666666667,
7
+ "fallback_rate": 0.5333333333333333,
8
+ "hidden_chart_oracle_success": 1.0,
9
+ "hidden_chart_oracle_utility": 2.0,
10
+ "outcome_ptr": 0.5333333333333333,
11
+ "predicted_margin": 0.27050518877972946,
12
+ "proposal_oracle_success": 0.6,
13
+ "proposal_oracle_utility": 1.4122399787108104,
14
+ "selected_candidate_index": 7.266666666666667,
15
+ "selected_success": 0.06666666666666667,
16
+ "selected_success_gain_over_base": -0.13333333333333333,
17
+ "selected_utility": 0.7441632241010666,
18
+ "selected_utility_gain_over_base": -0.20570731461048125,
19
+ "selector_regret": 0.6957427720228831,
20
+ "success_selector_gap": 0.5333333333333333,
21
+ "success_support_gap": 0.4,
22
+ "support_gap": 0.5877600212891897,
23
+ "tau": 0.2903110400293744
24
+ },
25
+ "PickCube-v1": {
26
+ "base_success": 0.0,
27
+ "base_utility": 0.5839591026306152,
28
+ "coverage": 0.0,
29
+ "execute_generated": 0.0,
30
+ "fallback_rate": 1.0,
31
+ "hidden_chart_oracle_success": 1.0,
32
+ "hidden_chart_oracle_utility": 2.0,
33
+ "outcome_ptr": 0.0,
34
+ "predicted_margin": 0.07614157399737802,
35
+ "proposal_oracle_success": 0.0,
36
+ "proposal_oracle_utility": 0.13425676027933756,
37
+ "selected_candidate_index": 7.333333333333333,
38
+ "selected_success": 0.0,
39
+ "selected_success_gain_over_base": 0.0,
40
+ "selected_utility": 0.5839591026306152,
41
+ "selected_utility_gain_over_base": 0.0,
42
+ "selector_regret": 0.0,
43
+ "success_selector_gap": 0.0,
44
+ "success_support_gap": 1.0,
45
+ "support_gap": 1.8657432397206624,
46
+ "tau": 0.29031104002937436
47
+ },
48
+ "PullCube-v1": {
49
+ "base_success": 0.36,
50
+ "base_utility": 0.7751704612234608,
51
+ "coverage": 0.5733333333333334,
52
+ "execute_generated": 0.5733333333333334,
53
+ "fallback_rate": 0.4266666666666667,
54
+ "hidden_chart_oracle_success": 0.52,
55
+ "hidden_chart_oracle_utility": 1.1467878860235214,
56
+ "outcome_ptr": 0.4666666666666667,
57
+ "predicted_margin": 0.3693117299280415,
58
+ "proposal_oracle_success": 0.5333333333333333,
59
+ "proposal_oracle_utility": 1.1093241602430741,
60
+ "selected_candidate_index": 4.266666666666667,
61
+ "selected_success": 0.3466666666666667,
62
+ "selected_success_gain_over_base": -0.013333333333333334,
63
+ "selected_utility": 0.7376457962176453,
64
+ "selected_utility_gain_over_base": -0.03752466500581553,
65
+ "selector_regret": 0.4480306339729577,
66
+ "success_selector_gap": 0.22666666666666666,
67
+ "success_support_gap": 0.02666666666666667,
68
+ "support_gap": 0.1075624164690574,
69
+ "tau": 0.29031104002937397
70
+ },
71
+ "PushCube-v1": {
72
+ "base_success": 0.16666666666666666,
73
+ "base_utility": 0.3924041787783305,
74
+ "coverage": 0.8333333333333334,
75
+ "execute_generated": 0.8333333333333334,
76
+ "fallback_rate": 0.16666666666666666,
77
+ "hidden_chart_oracle_success": 0.8333333333333334,
78
+ "hidden_chart_oracle_utility": 1.7015800600250561,
79
+ "outcome_ptr": 0.8333333333333334,
80
+ "predicted_margin": 0.4634462366336652,
81
+ "proposal_oracle_success": 0.8888888888888888,
82
+ "proposal_oracle_utility": 1.7857814149724112,
83
+ "selected_candidate_index": 7.055555555555555,
84
+ "selected_success": 0.2222222222222222,
85
+ "selected_success_gain_over_base": 0.05555555555555555,
86
+ "selected_utility": 0.5195176344778802,
87
+ "selected_utility_gain_over_base": 0.1271134556995498,
88
+ "selector_regret": 1.266263780494531,
89
+ "success_selector_gap": 0.6666666666666666,
90
+ "success_support_gap": 0.1111111111111111,
91
+ "support_gap": 0.21421858502758873,
92
+ "tau": 0.2903110400293744
93
+ },
94
+ "StackCube-v1": {
95
+ "base_success": 0.2727272727272727,
96
+ "base_utility": 0.7763947045261209,
97
+ "coverage": 0.3333333333333333,
98
+ "execute_generated": 0.3333333333333333,
99
+ "fallback_rate": 0.6666666666666666,
100
+ "hidden_chart_oracle_success": 1.0,
101
+ "hidden_chart_oracle_utility": 2.0,
102
+ "outcome_ptr": 0.6363636363636364,
103
+ "predicted_margin": 0.29484464535625693,
104
+ "proposal_oracle_success": 0.5151515151515151,
105
+ "proposal_oracle_utility": 1.1567951707225856,
106
+ "selected_candidate_index": 6.606060606060606,
107
+ "selected_success": 0.21212121212121213,
108
+ "selected_success_gain_over_base": -0.06060606060606061,
109
+ "selected_utility": 0.6815915500575845,
110
+ "selected_utility_gain_over_base": -0.09480315446853638,
111
+ "selector_regret": 0.4787966058109746,
112
+ "success_selector_gap": 0.30303030303030304,
113
+ "success_support_gap": 0.48484848484848486,
114
+ "support_gap": 0.8432048292774142,
115
+ "tau": 0.2903110400293744
116
+ }
117
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/report.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Nonlinear Train-Calibrated CTT Selector
2
+
3
+ Calibration rows: `432`
4
+ Fit rows: `277`
5
+ Selection rows: `155`
6
+ Eval rows: `144`
7
+ Selected model: `hgb_classifier`
8
+ Selected params: `{"l2_regularization": 0.01, "learning_rate": 0.05, "max_iter": 80, "max_leaf_nodes": 15, "random_state": 0}`
9
+ Tau: `0.290311`
10
+ Feature set: `chart_source_compat`
11
+ Target: `utility_margin`
12
+
13
+ The model is fit on calibration-fit rows, and model/tau are selected on held-out calibration-selection rows only. Eval outcomes are used only for reporting.
14
+
15
+ | Split | Coverage | Fallback | Base success | Selected success | Proposal oracle | OutcomePTR | Success support gap | Success selector gap |
16
+ | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
17
+ | fit | 0.4982 | 0.5018 | 0.2780 | 0.4657 | 0.5343 | 0.5343 | 0.1733 | 0.1191 |
18
+ | selection | 0.6129 | 0.3871 | 0.2774 | 0.4065 | 0.5419 | 0.5677 | 0.1806 | 0.1677 |
19
+ | eval | 0.5278 | 0.4722 | 0.2917 | 0.2639 | 0.5694 | 0.5486 | 0.2014 | 0.3264 |
20
+
21
+ This is a train-calibrated selector diagnostic over already measured candidates, not a new rollout.
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/split_hash.txt ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "calibration_target_index": {
3
+ "content_hash": "67f7b4a692f7c7e71da24378dc71e3399c5d35afc14c494869eba6087b765c42",
4
+ "retrieval_index_allowed": true,
5
+ "split": "train",
6
+ "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372"
7
+ },
8
+ "eval_target_index": {
9
+ "content_hash": "0c38d73dc9bc4ab756e12006671b5d12f1a23358b281e54b3f6a9ae3cbf6e9af",
10
+ "retrieval_index_allowed": false,
11
+ "split": "test",
12
+ "split_hash": "84151df9e88c219b76751677558cc5bbbc19082bf358cea7330158e9e403e25f"
13
+ },
14
+ "source_index": {
15
+ "content_hash": "67f7b4a692f7c7e71da24378dc71e3399c5d35afc14c494869eba6087b765c42",
16
+ "retrieval_index_allowed": true,
17
+ "split": "train",
18
+ "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372"
19
+ }
20
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/table.tex ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ % Auto-generated by scripts/eval_nonlinear_dominance_selector.py
2
+ \begin{tabular}{lrrrrrrrr}
3
+ \toprule
4
+ Rows & Coverage & Fallback & Base succ. & Selected succ. & Oracle succ. & OutcomePTR & Succ. support gap & Succ. selector gap \\
5
+ \midrule
6
+ 144 & 0.5278 & 0.4722 & 0.2917 & 0.2639 & 0.5694 & 0.5486 & 0.2014 & 0.3264 \\
7
+ \bottomrule
8
+ \end{tabular}
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chart_sourcecompat_obs_utility_envclip_k16_train_to_test/train.log ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ trained nonlinear dominance selector on calibration-fit rows only
2
+ selected_model_type=hgb_classifier
3
+ selected_model_params={"l2_regularization": 0.01, "learning_rate": 0.05, "max_iter": 80, "max_leaf_nodes": 15, "random_state": 0}
4
+ tau=0.290311
5
+ calibration_input=runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/command.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ python scripts/eval_nonlinear_dominance_selector.py --calibration-input runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json --calibration-target-index data/cil_charts_rgb_refs/train/index.json --eval-input runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json --eval-target-index data/cil_charts_rgb_refs/test/index.json --source-index data/cil_charts_rgb_refs/train/index.json --checkpoint-template runs/ctt_residual_base_context_obs_seed{seed}/model.pt --out-dir runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test --feature-set chart_compat --selector-chart-feature-mode base_context_obs --target positive_margin --k 16 --bootstrap-samples 1000
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/config.yaml ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ bootstrap_samples: 1000
2
+ calibration_input: runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json
3
+ calibration_target_index: data/cil_charts_rgb_refs/train/index.json
4
+ checkpoint_template: runs/ctt_residual_base_context_obs_seed{seed}/model.pt
5
+ eval_input: runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json
6
+ eval_target_index: data/cil_charts_rgb_refs/test/index.json
7
+ feature_set: chart_compat
8
+ k: 16
9
+ model_types: hgb_classifier,hgb_regressor,rf_regressor
10
+ out_dir: runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test
11
+ seed: 0
12
+ selection_frac: 0.35
13
+ selector_chart_feature_mode: base_context_obs
14
+ source_index: data/cil_charts_rgb_refs/train/index.json
15
+ target: positive_margin
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/data_hash.txt ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "calibration_input": "d79e275b896eb98caa128470d4614b76960cab12407faa157c583224be7abe80",
3
+ "calibration_target_index": "67973c50072aa87a3b97533b6292b08a8b4d88a95571dc22e91b6e8fd19e0d85",
4
+ "eval_input": "541d5a928da5f7ee04ca3dea570e04208830dffde431edfe5b5e38b692634045",
5
+ "eval_target_index": "dee1b5225485df6624747dd4c5074bcf514685b54017d05e698d2051638fbf36",
6
+ "source_index": "67973c50072aa87a3b97533b6292b08a8b4d88a95571dc22e91b6e8fd19e0d85"
7
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/eval.log ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ selected model/tau on held-out calibration-selection rows; evaluated on held-out measured rollout rows
2
+ eval_input=runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json
3
+ num_eval_rows=144
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/git_hash.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ 14e81292e2004925f81e0bb4810f6f144bfd87b7
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/metrics.json ADDED
The diff for this file is too large to render. See raw diff
 
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/metrics_by_seed.json ADDED
@@ -0,0 +1,2762 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "1073161778/train0": {
3
+ "base_success": 1.0,
4
+ "base_utility": 2.0,
5
+ "coverage": 0.0,
6
+ "execute_generated": 0.0,
7
+ "fallback_rate": 1.0,
8
+ "hidden_chart_oracle_success": 1.0,
9
+ "hidden_chart_oracle_utility": 2.0,
10
+ "outcome_ptr": 0.0,
11
+ "predicted_margin": 0.28355827608242967,
12
+ "proposal_oracle_success": 1.0,
13
+ "proposal_oracle_utility": 2.0,
14
+ "selected_candidate_index": 6.0,
15
+ "selected_success": 1.0,
16
+ "selected_success_gain_over_base": 0.0,
17
+ "selected_utility": 2.0,
18
+ "selected_utility_gain_over_base": 0.0,
19
+ "selector_regret": 0.0,
20
+ "success_selector_gap": 0.0,
21
+ "success_support_gap": 0.0,
22
+ "support_gap": 0.0,
23
+ "tau": 0.3473188188503786
24
+ },
25
+ "1073161778/train1": {
26
+ "base_success": 1.0,
27
+ "base_utility": 2.0,
28
+ "coverage": 0.0,
29
+ "execute_generated": 0.0,
30
+ "fallback_rate": 1.0,
31
+ "hidden_chart_oracle_success": 1.0,
32
+ "hidden_chart_oracle_utility": 2.0,
33
+ "outcome_ptr": 0.0,
34
+ "predicted_margin": 0.3337636013451598,
35
+ "proposal_oracle_success": 1.0,
36
+ "proposal_oracle_utility": 2.0,
37
+ "selected_candidate_index": 7.0,
38
+ "selected_success": 1.0,
39
+ "selected_success_gain_over_base": 0.0,
40
+ "selected_utility": 2.0,
41
+ "selected_utility_gain_over_base": 0.0,
42
+ "selector_regret": 0.0,
43
+ "success_selector_gap": 0.0,
44
+ "success_support_gap": 0.0,
45
+ "support_gap": 0.0,
46
+ "tau": 0.3473188188503786
47
+ },
48
+ "1073161778/train2": {
49
+ "base_success": 1.0,
50
+ "base_utility": 2.0,
51
+ "coverage": 0.0,
52
+ "execute_generated": 0.0,
53
+ "fallback_rate": 1.0,
54
+ "hidden_chart_oracle_success": 1.0,
55
+ "hidden_chart_oracle_utility": 2.0,
56
+ "outcome_ptr": 0.0,
57
+ "predicted_margin": 0.29522744379276233,
58
+ "proposal_oracle_success": 1.0,
59
+ "proposal_oracle_utility": 2.0,
60
+ "selected_candidate_index": 3.0,
61
+ "selected_success": 1.0,
62
+ "selected_success_gain_over_base": 0.0,
63
+ "selected_utility": 2.0,
64
+ "selected_utility_gain_over_base": 0.0,
65
+ "selector_regret": 0.0,
66
+ "success_selector_gap": 0.0,
67
+ "success_support_gap": 0.0,
68
+ "support_gap": 0.0,
69
+ "tau": 0.3473188188503786
70
+ },
71
+ "1113371261/train0": {
72
+ "base_success": 0.0,
73
+ "base_utility": 0.15842342376708984,
74
+ "coverage": 0.0,
75
+ "execute_generated": 0.0,
76
+ "fallback_rate": 1.0,
77
+ "hidden_chart_oracle_success": 1.0,
78
+ "hidden_chart_oracle_utility": 2.0,
79
+ "outcome_ptr": 0.0,
80
+ "predicted_margin": 0.32662525863909614,
81
+ "proposal_oracle_success": 0.0,
82
+ "proposal_oracle_utility": 0.10274861752986908,
83
+ "selected_candidate_index": 0.0,
84
+ "selected_success": 0.0,
85
+ "selected_success_gain_over_base": 0.0,
86
+ "selected_utility": 0.15842342376708984,
87
+ "selected_utility_gain_over_base": 0.0,
88
+ "selector_regret": 0.0,
89
+ "success_selector_gap": 0.0,
90
+ "success_support_gap": 1.0,
91
+ "support_gap": 1.897251382470131,
92
+ "tau": 0.3473188188503786
93
+ },
94
+ "1113371261/train1": {
95
+ "base_success": 0.0,
96
+ "base_utility": 0.15842342376708984,
97
+ "coverage": 0.0,
98
+ "execute_generated": 0.0,
99
+ "fallback_rate": 1.0,
100
+ "hidden_chart_oracle_success": 1.0,
101
+ "hidden_chart_oracle_utility": 2.0,
102
+ "outcome_ptr": 1.0,
103
+ "predicted_margin": 0.27481853311866083,
104
+ "proposal_oracle_success": 0.0,
105
+ "proposal_oracle_utility": 0.18093222379684448,
106
+ "selected_candidate_index": 4.0,
107
+ "selected_success": 0.0,
108
+ "selected_success_gain_over_base": 0.0,
109
+ "selected_utility": 0.15842342376708984,
110
+ "selected_utility_gain_over_base": 0.0,
111
+ "selector_regret": 0.02250880002975464,
112
+ "success_selector_gap": 0.0,
113
+ "success_support_gap": 1.0,
114
+ "support_gap": 1.8190677762031555,
115
+ "tau": 0.3473188188503786
116
+ },
117
+ "1113371261/train2": {
118
+ "base_success": 0.0,
119
+ "base_utility": 0.15842342376708984,
120
+ "coverage": 0.0,
121
+ "execute_generated": 0.0,
122
+ "fallback_rate": 1.0,
123
+ "hidden_chart_oracle_success": 1.0,
124
+ "hidden_chart_oracle_utility": 2.0,
125
+ "outcome_ptr": 1.0,
126
+ "predicted_margin": 0.3131801253335776,
127
+ "proposal_oracle_success": 0.0,
128
+ "proposal_oracle_utility": 0.19527167081832886,
129
+ "selected_candidate_index": 2.0,
130
+ "selected_success": 0.0,
131
+ "selected_success_gain_over_base": 0.0,
132
+ "selected_utility": 0.15842342376708984,
133
+ "selected_utility_gain_over_base": 0.0,
134
+ "selector_regret": 0.036848247051239014,
135
+ "success_selector_gap": 0.0,
136
+ "success_support_gap": 1.0,
137
+ "support_gap": 1.8047283291816711,
138
+ "tau": 0.3473188188503786
139
+ },
140
+ "1151410346/train0": {
141
+ "base_success": 0.0,
142
+ "base_utility": 0.2260012924671173,
143
+ "coverage": 0.0,
144
+ "execute_generated": 0.0,
145
+ "fallback_rate": 1.0,
146
+ "hidden_chart_oracle_success": 1.0,
147
+ "hidden_chart_oracle_utility": 2.0,
148
+ "outcome_ptr": 1.0,
149
+ "predicted_margin": 0.20532758727946646,
150
+ "proposal_oracle_success": 0.0,
151
+ "proposal_oracle_utility": 0.8663521409034729,
152
+ "selected_candidate_index": 7.0,
153
+ "selected_success": 0.0,
154
+ "selected_success_gain_over_base": 0.0,
155
+ "selected_utility": 0.2260012924671173,
156
+ "selected_utility_gain_over_base": 0.0,
157
+ "selector_regret": 0.6403508484363556,
158
+ "success_selector_gap": 0.0,
159
+ "success_support_gap": 1.0,
160
+ "support_gap": 1.133647859096527,
161
+ "tau": 0.3473188188503786
162
+ },
163
+ "1151410346/train1": {
164
+ "base_success": 0.0,
165
+ "base_utility": 0.2260012924671173,
166
+ "coverage": 0.0,
167
+ "execute_generated": 0.0,
168
+ "fallback_rate": 1.0,
169
+ "hidden_chart_oracle_success": 1.0,
170
+ "hidden_chart_oracle_utility": 2.0,
171
+ "outcome_ptr": 1.0,
172
+ "predicted_margin": 0.21665335322366186,
173
+ "proposal_oracle_success": 1.0,
174
+ "proposal_oracle_utility": 2.0,
175
+ "selected_candidate_index": 9.0,
176
+ "selected_success": 0.0,
177
+ "selected_success_gain_over_base": 0.0,
178
+ "selected_utility": 0.2260012924671173,
179
+ "selected_utility_gain_over_base": 0.0,
180
+ "selector_regret": 1.7739987075328827,
181
+ "success_selector_gap": 1.0,
182
+ "success_support_gap": 0.0,
183
+ "support_gap": 0.0,
184
+ "tau": 0.3473188188503786
185
+ },
186
+ "1151410346/train2": {
187
+ "base_success": 0.0,
188
+ "base_utility": 0.2260012924671173,
189
+ "coverage": 0.0,
190
+ "execute_generated": 0.0,
191
+ "fallback_rate": 1.0,
192
+ "hidden_chart_oracle_success": 1.0,
193
+ "hidden_chart_oracle_utility": 2.0,
194
+ "outcome_ptr": 1.0,
195
+ "predicted_margin": 0.22061568738801626,
196
+ "proposal_oracle_success": 0.0,
197
+ "proposal_oracle_utility": 0.5522047281265259,
198
+ "selected_candidate_index": 10.0,
199
+ "selected_success": 0.0,
200
+ "selected_success_gain_over_base": 0.0,
201
+ "selected_utility": 0.2260012924671173,
202
+ "selected_utility_gain_over_base": 0.0,
203
+ "selector_regret": 0.32620343565940857,
204
+ "success_selector_gap": 0.0,
205
+ "success_support_gap": 1.0,
206
+ "support_gap": 1.4477952718734741,
207
+ "tau": 0.3473188188503786
208
+ },
209
+ "1261930456/train0": {
210
+ "base_success": 1.0,
211
+ "base_utility": 2.0,
212
+ "coverage": 1.0,
213
+ "execute_generated": 1.0,
214
+ "fallback_rate": 0.0,
215
+ "hidden_chart_oracle_success": 1.0,
216
+ "hidden_chart_oracle_utility": 2.0,
217
+ "outcome_ptr": 0.0,
218
+ "predicted_margin": 0.5566373654374286,
219
+ "proposal_oracle_success": 1.0,
220
+ "proposal_oracle_utility": 2.0,
221
+ "selected_candidate_index": 5.0,
222
+ "selected_success": 1.0,
223
+ "selected_success_gain_over_base": 0.0,
224
+ "selected_utility": 2.0,
225
+ "selected_utility_gain_over_base": 0.0,
226
+ "selector_regret": 0.0,
227
+ "success_selector_gap": 0.0,
228
+ "success_support_gap": 0.0,
229
+ "support_gap": 0.0,
230
+ "tau": 0.3473188188503786
231
+ },
232
+ "1261930456/train1": {
233
+ "base_success": 1.0,
234
+ "base_utility": 2.0,
235
+ "coverage": 1.0,
236
+ "execute_generated": 1.0,
237
+ "fallback_rate": 0.0,
238
+ "hidden_chart_oracle_success": 1.0,
239
+ "hidden_chart_oracle_utility": 2.0,
240
+ "outcome_ptr": 0.0,
241
+ "predicted_margin": 0.5940825716503084,
242
+ "proposal_oracle_success": 1.0,
243
+ "proposal_oracle_utility": 2.0,
244
+ "selected_candidate_index": 0.0,
245
+ "selected_success": 1.0,
246
+ "selected_success_gain_over_base": 0.0,
247
+ "selected_utility": 2.0,
248
+ "selected_utility_gain_over_base": 0.0,
249
+ "selector_regret": 0.0,
250
+ "success_selector_gap": 0.0,
251
+ "success_support_gap": 0.0,
252
+ "support_gap": 0.0,
253
+ "tau": 0.3473188188503786
254
+ },
255
+ "1261930456/train2": {
256
+ "base_success": 1.0,
257
+ "base_utility": 2.0,
258
+ "coverage": 1.0,
259
+ "execute_generated": 1.0,
260
+ "fallback_rate": 0.0,
261
+ "hidden_chart_oracle_success": 1.0,
262
+ "hidden_chart_oracle_utility": 2.0,
263
+ "outcome_ptr": 0.0,
264
+ "predicted_margin": 0.5877928871421093,
265
+ "proposal_oracle_success": 1.0,
266
+ "proposal_oracle_utility": 2.0,
267
+ "selected_candidate_index": 1.0,
268
+ "selected_success": 0.0,
269
+ "selected_success_gain_over_base": -1.0,
270
+ "selected_utility": 0.04538704454898834,
271
+ "selected_utility_gain_over_base": -1.9546129554510117,
272
+ "selector_regret": 1.9546129554510117,
273
+ "success_selector_gap": 1.0,
274
+ "success_support_gap": 0.0,
275
+ "support_gap": 0.0,
276
+ "tau": 0.3473188188503786
277
+ },
278
+ "1324133122/train0": {
279
+ "base_success": 0.0,
280
+ "base_utility": 0.09220491349697113,
281
+ "coverage": 0.0,
282
+ "execute_generated": 0.0,
283
+ "fallback_rate": 1.0,
284
+ "hidden_chart_oracle_success": 1.0,
285
+ "hidden_chart_oracle_utility": 2.0,
286
+ "outcome_ptr": 0.0,
287
+ "predicted_margin": 0.1953526650841092,
288
+ "proposal_oracle_success": 0.0,
289
+ "proposal_oracle_utility": 0.054989203810691833,
290
+ "selected_candidate_index": 0.0,
291
+ "selected_success": 0.0,
292
+ "selected_success_gain_over_base": 0.0,
293
+ "selected_utility": 0.09220491349697113,
294
+ "selected_utility_gain_over_base": 0.0,
295
+ "selector_regret": 0.0,
296
+ "success_selector_gap": 0.0,
297
+ "success_support_gap": 1.0,
298
+ "support_gap": 1.9450107961893082,
299
+ "tau": 0.3473188188503786
300
+ },
301
+ "1324133122/train1": {
302
+ "base_success": 0.0,
303
+ "base_utility": 0.09220491349697113,
304
+ "coverage": 0.0,
305
+ "execute_generated": 0.0,
306
+ "fallback_rate": 1.0,
307
+ "hidden_chart_oracle_success": 1.0,
308
+ "hidden_chart_oracle_utility": 2.0,
309
+ "outcome_ptr": 1.0,
310
+ "predicted_margin": 0.10311308876908776,
311
+ "proposal_oracle_success": 0.0,
312
+ "proposal_oracle_utility": 0.1324586570262909,
313
+ "selected_candidate_index": 5.0,
314
+ "selected_success": 0.0,
315
+ "selected_success_gain_over_base": 0.0,
316
+ "selected_utility": 0.09220491349697113,
317
+ "selected_utility_gain_over_base": 0.0,
318
+ "selector_regret": 0.04025374352931976,
319
+ "success_selector_gap": 0.0,
320
+ "success_support_gap": 1.0,
321
+ "support_gap": 1.867541342973709,
322
+ "tau": 0.3473188188503786
323
+ },
324
+ "1324133122/train2": {
325
+ "base_success": 0.0,
326
+ "base_utility": 0.09220491349697113,
327
+ "coverage": 0.0,
328
+ "execute_generated": 0.0,
329
+ "fallback_rate": 1.0,
330
+ "hidden_chart_oracle_success": 1.0,
331
+ "hidden_chart_oracle_utility": 2.0,
332
+ "outcome_ptr": 1.0,
333
+ "predicted_margin": 0.17894998490871183,
334
+ "proposal_oracle_success": 0.0,
335
+ "proposal_oracle_utility": 0.19291165471076965,
336
+ "selected_candidate_index": 7.0,
337
+ "selected_success": 0.0,
338
+ "selected_success_gain_over_base": 0.0,
339
+ "selected_utility": 0.09220491349697113,
340
+ "selected_utility_gain_over_base": 0.0,
341
+ "selector_regret": 0.10070674121379852,
342
+ "success_selector_gap": 0.0,
343
+ "success_support_gap": 1.0,
344
+ "support_gap": 1.8070883452892303,
345
+ "tau": 0.3473188188503786
346
+ },
347
+ "1324642411/train0": {
348
+ "base_success": 0.0,
349
+ "base_utility": 0.10647052526473999,
350
+ "coverage": 1.0,
351
+ "execute_generated": 1.0,
352
+ "fallback_rate": 0.0,
353
+ "hidden_chart_oracle_success": 1.0,
354
+ "hidden_chart_oracle_utility": 2.0,
355
+ "outcome_ptr": 1.0,
356
+ "predicted_margin": 0.5244761772076728,
357
+ "proposal_oracle_success": 0.0,
358
+ "proposal_oracle_utility": 0.14406004548072815,
359
+ "selected_candidate_index": 3.0,
360
+ "selected_success": 0.0,
361
+ "selected_success_gain_over_base": 0.0,
362
+ "selected_utility": 0.061532557010650635,
363
+ "selected_utility_gain_over_base": -0.044937968254089355,
364
+ "selector_regret": 0.08252748847007751,
365
+ "success_selector_gap": 0.0,
366
+ "success_support_gap": 1.0,
367
+ "support_gap": 1.8559399545192719,
368
+ "tau": 0.3473188188503786
369
+ },
370
+ "1324642411/train1": {
371
+ "base_success": 0.0,
372
+ "base_utility": 0.10647052526473999,
373
+ "coverage": 1.0,
374
+ "execute_generated": 1.0,
375
+ "fallback_rate": 0.0,
376
+ "hidden_chart_oracle_success": 1.0,
377
+ "hidden_chart_oracle_utility": 2.0,
378
+ "outcome_ptr": 1.0,
379
+ "predicted_margin": 0.48070996484280953,
380
+ "proposal_oracle_success": 1.0,
381
+ "proposal_oracle_utility": 2.0,
382
+ "selected_candidate_index": 9.0,
383
+ "selected_success": 0.0,
384
+ "selected_success_gain_over_base": 0.0,
385
+ "selected_utility": 0.1280023753643036,
386
+ "selected_utility_gain_over_base": 0.0215318500995636,
387
+ "selector_regret": 1.8719976246356964,
388
+ "success_selector_gap": 1.0,
389
+ "success_support_gap": 0.0,
390
+ "support_gap": 0.0,
391
+ "tau": 0.3473188188503786
392
+ },
393
+ "1324642411/train2": {
394
+ "base_success": 0.0,
395
+ "base_utility": 0.10647052526473999,
396
+ "coverage": 1.0,
397
+ "execute_generated": 1.0,
398
+ "fallback_rate": 0.0,
399
+ "hidden_chart_oracle_success": 1.0,
400
+ "hidden_chart_oracle_utility": 2.0,
401
+ "outcome_ptr": 1.0,
402
+ "predicted_margin": 0.4884411626074505,
403
+ "proposal_oracle_success": 1.0,
404
+ "proposal_oracle_utility": 2.0,
405
+ "selected_candidate_index": 3.0,
406
+ "selected_success": 1.0,
407
+ "selected_success_gain_over_base": 1.0,
408
+ "selected_utility": 2.0,
409
+ "selected_utility_gain_over_base": 1.89352947473526,
410
+ "selector_regret": 0.0,
411
+ "success_selector_gap": 0.0,
412
+ "success_support_gap": 0.0,
413
+ "support_gap": 0.0,
414
+ "tau": 0.3473188188503786
415
+ },
416
+ "1375492786/train0": {
417
+ "base_success": 1.0,
418
+ "base_utility": 2.0,
419
+ "coverage": 0.0,
420
+ "execute_generated": 0.0,
421
+ "fallback_rate": 1.0,
422
+ "hidden_chart_oracle_success": 1.0,
423
+ "hidden_chart_oracle_utility": 2.0,
424
+ "outcome_ptr": 0.0,
425
+ "predicted_margin": 0.16043123876023302,
426
+ "proposal_oracle_success": 1.0,
427
+ "proposal_oracle_utility": 2.0,
428
+ "selected_candidate_index": 15.0,
429
+ "selected_success": 1.0,
430
+ "selected_success_gain_over_base": 0.0,
431
+ "selected_utility": 2.0,
432
+ "selected_utility_gain_over_base": 0.0,
433
+ "selector_regret": 0.0,
434
+ "success_selector_gap": 0.0,
435
+ "success_support_gap": 0.0,
436
+ "support_gap": 0.0,
437
+ "tau": 0.3473188188503786
438
+ },
439
+ "1375492786/train1": {
440
+ "base_success": 1.0,
441
+ "base_utility": 2.0,
442
+ "coverage": 0.0,
443
+ "execute_generated": 0.0,
444
+ "fallback_rate": 1.0,
445
+ "hidden_chart_oracle_success": 1.0,
446
+ "hidden_chart_oracle_utility": 2.0,
447
+ "outcome_ptr": 0.0,
448
+ "predicted_margin": 0.17011388186485987,
449
+ "proposal_oracle_success": 1.0,
450
+ "proposal_oracle_utility": 2.0,
451
+ "selected_candidate_index": 10.0,
452
+ "selected_success": 1.0,
453
+ "selected_success_gain_over_base": 0.0,
454
+ "selected_utility": 2.0,
455
+ "selected_utility_gain_over_base": 0.0,
456
+ "selector_regret": 0.0,
457
+ "success_selector_gap": 0.0,
458
+ "success_support_gap": 0.0,
459
+ "support_gap": 0.0,
460
+ "tau": 0.3473188188503786
461
+ },
462
+ "1375492786/train2": {
463
+ "base_success": 1.0,
464
+ "base_utility": 2.0,
465
+ "coverage": 0.0,
466
+ "execute_generated": 0.0,
467
+ "fallback_rate": 1.0,
468
+ "hidden_chart_oracle_success": 1.0,
469
+ "hidden_chart_oracle_utility": 2.0,
470
+ "outcome_ptr": 0.0,
471
+ "predicted_margin": 0.15542103752775793,
472
+ "proposal_oracle_success": 1.0,
473
+ "proposal_oracle_utility": 2.0,
474
+ "selected_candidate_index": 15.0,
475
+ "selected_success": 1.0,
476
+ "selected_success_gain_over_base": 0.0,
477
+ "selected_utility": 2.0,
478
+ "selected_utility_gain_over_base": 0.0,
479
+ "selector_regret": 0.0,
480
+ "success_selector_gap": 0.0,
481
+ "success_support_gap": 0.0,
482
+ "support_gap": 0.0,
483
+ "tau": 0.3473188188503786
484
+ },
485
+ "141984615/train0": {
486
+ "base_success": 0.0,
487
+ "base_utility": 0.10283424705266953,
488
+ "coverage": 1.0,
489
+ "execute_generated": 1.0,
490
+ "fallback_rate": 0.0,
491
+ "hidden_chart_oracle_success": 0.0,
492
+ "hidden_chart_oracle_utility": 0.2857772707939148,
493
+ "outcome_ptr": 0.0,
494
+ "predicted_margin": 0.3988789611600871,
495
+ "proposal_oracle_success": 0.0,
496
+ "proposal_oracle_utility": 0.015177905559539795,
497
+ "selected_candidate_index": 1.0,
498
+ "selected_success": 0.0,
499
+ "selected_success_gain_over_base": 0.0,
500
+ "selected_utility": 0.0042244791984558105,
501
+ "selected_utility_gain_over_base": -0.09860976785421371,
502
+ "selector_regret": 0.010953426361083984,
503
+ "success_selector_gap": 0.0,
504
+ "success_support_gap": 0.0,
505
+ "support_gap": 0.270599365234375,
506
+ "tau": 0.3473188188503786
507
+ },
508
+ "141984615/train1": {
509
+ "base_success": 0.0,
510
+ "base_utility": 0.10283424705266953,
511
+ "coverage": 1.0,
512
+ "execute_generated": 1.0,
513
+ "fallback_rate": 0.0,
514
+ "hidden_chart_oracle_success": 0.0,
515
+ "hidden_chart_oracle_utility": 0.2857772707939148,
516
+ "outcome_ptr": 0.0,
517
+ "predicted_margin": 0.37538333059509166,
518
+ "proposal_oracle_success": 0.0,
519
+ "proposal_oracle_utility": 0.037011127918958664,
520
+ "selected_candidate_index": 6.0,
521
+ "selected_success": 0.0,
522
+ "selected_success_gain_over_base": 0.0,
523
+ "selected_utility": 0.035802166908979416,
524
+ "selected_utility_gain_over_base": -0.06703208014369011,
525
+ "selector_regret": 0.001208961009979248,
526
+ "success_selector_gap": 0.0,
527
+ "success_support_gap": 0.0,
528
+ "support_gap": 0.24876614287495613,
529
+ "tau": 0.3473188188503786
530
+ },
531
+ "141984615/train2": {
532
+ "base_success": 0.0,
533
+ "base_utility": 0.10283424705266953,
534
+ "coverage": 0.0,
535
+ "execute_generated": 0.0,
536
+ "fallback_rate": 1.0,
537
+ "hidden_chart_oracle_success": 0.0,
538
+ "hidden_chart_oracle_utility": 0.2857772707939148,
539
+ "outcome_ptr": 0.0,
540
+ "predicted_margin": 0.30448894670641713,
541
+ "proposal_oracle_success": 0.0,
542
+ "proposal_oracle_utility": 0.058951716870069504,
543
+ "selected_candidate_index": 4.0,
544
+ "selected_success": 0.0,
545
+ "selected_success_gain_over_base": 0.0,
546
+ "selected_utility": 0.10283424705266953,
547
+ "selected_utility_gain_over_base": 0.0,
548
+ "selector_regret": 0.0,
549
+ "success_selector_gap": 0.0,
550
+ "success_support_gap": 0.0,
551
+ "support_gap": 0.2268255539238453,
552
+ "tau": 0.3473188188503786
553
+ },
554
+ "1461069713/train0": {
555
+ "base_success": 0.0,
556
+ "base_utility": 0.0939030647277832,
557
+ "coverage": 0.0,
558
+ "execute_generated": 0.0,
559
+ "fallback_rate": 1.0,
560
+ "hidden_chart_oracle_success": 0.0,
561
+ "hidden_chart_oracle_utility": 0.1593424528837204,
562
+ "outcome_ptr": 0.0,
563
+ "predicted_margin": 0.3050981300072402,
564
+ "proposal_oracle_success": 0.0,
565
+ "proposal_oracle_utility": 0.06019836664199829,
566
+ "selected_candidate_index": 2.0,
567
+ "selected_success": 0.0,
568
+ "selected_success_gain_over_base": 0.0,
569
+ "selected_utility": 0.0939030647277832,
570
+ "selected_utility_gain_over_base": 0.0,
571
+ "selector_regret": 0.0,
572
+ "success_selector_gap": 0.0,
573
+ "success_support_gap": 0.0,
574
+ "support_gap": 0.0991440862417221,
575
+ "tau": 0.3473188188503786
576
+ },
577
+ "1461069713/train1": {
578
+ "base_success": 0.0,
579
+ "base_utility": 0.0939030647277832,
580
+ "coverage": 1.0,
581
+ "execute_generated": 1.0,
582
+ "fallback_rate": 0.0,
583
+ "hidden_chart_oracle_success": 0.0,
584
+ "hidden_chart_oracle_utility": 0.1593424528837204,
585
+ "outcome_ptr": 0.0,
586
+ "predicted_margin": 0.7484074094930177,
587
+ "proposal_oracle_success": 0.0,
588
+ "proposal_oracle_utility": 0.016585012897849083,
589
+ "selected_candidate_index": 0.0,
590
+ "selected_success": 0.0,
591
+ "selected_success_gain_over_base": 0.0,
592
+ "selected_utility": 0.009687979705631733,
593
+ "selected_utility_gain_over_base": -0.08421508502215147,
594
+ "selector_regret": 0.00689703319221735,
595
+ "success_selector_gap": 0.0,
596
+ "success_support_gap": 0.0,
597
+ "support_gap": 0.14275743998587132,
598
+ "tau": 0.3473188188503786
599
+ },
600
+ "1461069713/train2": {
601
+ "base_success": 0.0,
602
+ "base_utility": 0.0939030647277832,
603
+ "coverage": 0.0,
604
+ "execute_generated": 0.0,
605
+ "fallback_rate": 1.0,
606
+ "hidden_chart_oracle_success": 0.0,
607
+ "hidden_chart_oracle_utility": 0.1593424528837204,
608
+ "outcome_ptr": 1.0,
609
+ "predicted_margin": 0.2910706981901193,
610
+ "proposal_oracle_success": 0.0,
611
+ "proposal_oracle_utility": 0.09997648000717163,
612
+ "selected_candidate_index": 5.0,
613
+ "selected_success": 0.0,
614
+ "selected_success_gain_over_base": 0.0,
615
+ "selected_utility": 0.0939030647277832,
616
+ "selected_utility_gain_over_base": 0.0,
617
+ "selector_regret": 0.006073415279388428,
618
+ "success_selector_gap": 0.0,
619
+ "success_support_gap": 0.0,
620
+ "support_gap": 0.05936597287654877,
621
+ "tau": 0.3473188188503786
622
+ },
623
+ "1480856810/train0": {
624
+ "base_success": 0.0,
625
+ "base_utility": 0.08703407645225525,
626
+ "coverage": 0.0,
627
+ "execute_generated": 0.0,
628
+ "fallback_rate": 1.0,
629
+ "hidden_chart_oracle_success": 0.0,
630
+ "hidden_chart_oracle_utility": 0.20948036015033722,
631
+ "outcome_ptr": 1.0,
632
+ "predicted_margin": 0.29221027147425893,
633
+ "proposal_oracle_success": 1.0,
634
+ "proposal_oracle_utility": 2.0,
635
+ "selected_candidate_index": 2.0,
636
+ "selected_success": 0.0,
637
+ "selected_success_gain_over_base": 0.0,
638
+ "selected_utility": 0.08703407645225525,
639
+ "selected_utility_gain_over_base": 0.0,
640
+ "selector_regret": 1.9129659235477448,
641
+ "success_selector_gap": 1.0,
642
+ "success_support_gap": 0.0,
643
+ "support_gap": 0.0,
644
+ "tau": 0.3473188188503786
645
+ },
646
+ "1480856810/train1": {
647
+ "base_success": 0.0,
648
+ "base_utility": 0.08703407645225525,
649
+ "coverage": 1.0,
650
+ "execute_generated": 1.0,
651
+ "fallback_rate": 0.0,
652
+ "hidden_chart_oracle_success": 0.0,
653
+ "hidden_chart_oracle_utility": 0.20948036015033722,
654
+ "outcome_ptr": 1.0,
655
+ "predicted_margin": 0.43921922377953376,
656
+ "proposal_oracle_success": 1.0,
657
+ "proposal_oracle_utility": 2.0,
658
+ "selected_candidate_index": 13.0,
659
+ "selected_success": 0.0,
660
+ "selected_success_gain_over_base": 0.0,
661
+ "selected_utility": 0.20681259036064148,
662
+ "selected_utility_gain_over_base": 0.11977851390838623,
663
+ "selector_regret": 1.7931874096393585,
664
+ "success_selector_gap": 1.0,
665
+ "success_support_gap": 0.0,
666
+ "support_gap": 0.0,
667
+ "tau": 0.3473188188503786
668
+ },
669
+ "1480856810/train2": {
670
+ "base_success": 0.0,
671
+ "base_utility": 0.08703407645225525,
672
+ "coverage": 0.0,
673
+ "execute_generated": 0.0,
674
+ "fallback_rate": 1.0,
675
+ "hidden_chart_oracle_success": 0.0,
676
+ "hidden_chart_oracle_utility": 0.20948036015033722,
677
+ "outcome_ptr": 1.0,
678
+ "predicted_margin": 0.29221027147425893,
679
+ "proposal_oracle_success": 1.0,
680
+ "proposal_oracle_utility": 2.0,
681
+ "selected_candidate_index": 1.0,
682
+ "selected_success": 0.0,
683
+ "selected_success_gain_over_base": 0.0,
684
+ "selected_utility": 0.08703407645225525,
685
+ "selected_utility_gain_over_base": 0.0,
686
+ "selector_regret": 1.9129659235477448,
687
+ "success_selector_gap": 1.0,
688
+ "success_support_gap": 0.0,
689
+ "support_gap": 0.0,
690
+ "tau": 0.3473188188503786
691
+ },
692
+ "1509980664/train0": {
693
+ "base_success": 0.0,
694
+ "base_utility": 0.8334100246429443,
695
+ "coverage": 1.0,
696
+ "execute_generated": 1.0,
697
+ "fallback_rate": 0.0,
698
+ "hidden_chart_oracle_success": 1.0,
699
+ "hidden_chart_oracle_utility": 2.0,
700
+ "outcome_ptr": 1.0,
701
+ "predicted_margin": 0.5870819789979614,
702
+ "proposal_oracle_success": 1.0,
703
+ "proposal_oracle_utility": 2.0,
704
+ "selected_candidate_index": 10.0,
705
+ "selected_success": 1.0,
706
+ "selected_success_gain_over_base": 1.0,
707
+ "selected_utility": 2.0,
708
+ "selected_utility_gain_over_base": 1.1665899753570557,
709
+ "selector_regret": 0.0,
710
+ "success_selector_gap": 0.0,
711
+ "success_support_gap": 0.0,
712
+ "support_gap": 0.0,
713
+ "tau": 0.3473188188503786
714
+ },
715
+ "1509980664/train1": {
716
+ "base_success": 0.0,
717
+ "base_utility": 0.8334100246429443,
718
+ "coverage": 1.0,
719
+ "execute_generated": 1.0,
720
+ "fallback_rate": 0.0,
721
+ "hidden_chart_oracle_success": 1.0,
722
+ "hidden_chart_oracle_utility": 2.0,
723
+ "outcome_ptr": 1.0,
724
+ "predicted_margin": 0.6097239173416052,
725
+ "proposal_oracle_success": 1.0,
726
+ "proposal_oracle_utility": 2.0,
727
+ "selected_candidate_index": 4.0,
728
+ "selected_success": 0.0,
729
+ "selected_success_gain_over_base": 0.0,
730
+ "selected_utility": 0.21804393827915192,
731
+ "selected_utility_gain_over_base": -0.6153660863637924,
732
+ "selector_regret": 1.781956061720848,
733
+ "success_selector_gap": 1.0,
734
+ "success_support_gap": 0.0,
735
+ "support_gap": 0.0,
736
+ "tau": 0.3473188188503786
737
+ },
738
+ "1509980664/train2": {
739
+ "base_success": 0.0,
740
+ "base_utility": 0.8334100246429443,
741
+ "coverage": 1.0,
742
+ "execute_generated": 1.0,
743
+ "fallback_rate": 0.0,
744
+ "hidden_chart_oracle_success": 1.0,
745
+ "hidden_chart_oracle_utility": 2.0,
746
+ "outcome_ptr": 1.0,
747
+ "predicted_margin": 0.5974323487206175,
748
+ "proposal_oracle_success": 1.0,
749
+ "proposal_oracle_utility": 2.0,
750
+ "selected_candidate_index": 12.0,
751
+ "selected_success": 0.0,
752
+ "selected_success_gain_over_base": 0.0,
753
+ "selected_utility": 0.8189421892166138,
754
+ "selected_utility_gain_over_base": -0.014467835426330566,
755
+ "selector_regret": 1.1810578107833862,
756
+ "success_selector_gap": 1.0,
757
+ "success_support_gap": 0.0,
758
+ "support_gap": 0.0,
759
+ "tau": 0.3473188188503786
760
+ },
761
+ "1522272686/train0": {
762
+ "base_success": 0.0,
763
+ "base_utility": 0.5839591026306152,
764
+ "coverage": 0.0,
765
+ "execute_generated": 0.0,
766
+ "fallback_rate": 1.0,
767
+ "hidden_chart_oracle_success": 1.0,
768
+ "hidden_chart_oracle_utility": 2.0,
769
+ "outcome_ptr": 0.0,
770
+ "predicted_margin": 0.08863633481483887,
771
+ "proposal_oracle_success": 0.0,
772
+ "proposal_oracle_utility": 0.18325337767601013,
773
+ "selected_candidate_index": 8.0,
774
+ "selected_success": 0.0,
775
+ "selected_success_gain_over_base": 0.0,
776
+ "selected_utility": 0.5839591026306152,
777
+ "selected_utility_gain_over_base": 0.0,
778
+ "selector_regret": 0.0,
779
+ "success_selector_gap": 0.0,
780
+ "success_support_gap": 1.0,
781
+ "support_gap": 1.8167466223239899,
782
+ "tau": 0.3473188188503786
783
+ },
784
+ "1522272686/train1": {
785
+ "base_success": 0.0,
786
+ "base_utility": 0.5839591026306152,
787
+ "coverage": 0.0,
788
+ "execute_generated": 0.0,
789
+ "fallback_rate": 1.0,
790
+ "hidden_chart_oracle_success": 1.0,
791
+ "hidden_chart_oracle_utility": 2.0,
792
+ "outcome_ptr": 0.0,
793
+ "predicted_margin": 0.09154681464635614,
794
+ "proposal_oracle_success": 0.0,
795
+ "proposal_oracle_utility": 0.07014383375644684,
796
+ "selected_candidate_index": 7.0,
797
+ "selected_success": 0.0,
798
+ "selected_success_gain_over_base": 0.0,
799
+ "selected_utility": 0.5839591026306152,
800
+ "selected_utility_gain_over_base": 0.0,
801
+ "selector_regret": 0.0,
802
+ "success_selector_gap": 0.0,
803
+ "success_support_gap": 1.0,
804
+ "support_gap": 1.9298561662435532,
805
+ "tau": 0.3473188188503786
806
+ },
807
+ "1522272686/train2": {
808
+ "base_success": 0.0,
809
+ "base_utility": 0.5839591026306152,
810
+ "coverage": 0.0,
811
+ "execute_generated": 0.0,
812
+ "fallback_rate": 1.0,
813
+ "hidden_chart_oracle_success": 1.0,
814
+ "hidden_chart_oracle_utility": 2.0,
815
+ "outcome_ptr": 0.0,
816
+ "predicted_margin": 0.08595256684637552,
817
+ "proposal_oracle_success": 0.0,
818
+ "proposal_oracle_utility": 0.14937306940555573,
819
+ "selected_candidate_index": 8.0,
820
+ "selected_success": 0.0,
821
+ "selected_success_gain_over_base": 0.0,
822
+ "selected_utility": 0.5839591026306152,
823
+ "selected_utility_gain_over_base": 0.0,
824
+ "selector_regret": 0.0,
825
+ "success_selector_gap": 0.0,
826
+ "success_support_gap": 1.0,
827
+ "support_gap": 1.8506269305944443,
828
+ "tau": 0.3473188188503786
829
+ },
830
+ "1549324805/train0": {
831
+ "base_success": 0.0,
832
+ "base_utility": 0.10451483726501465,
833
+ "coverage": 1.0,
834
+ "execute_generated": 1.0,
835
+ "fallback_rate": 0.0,
836
+ "hidden_chart_oracle_success": 0.0,
837
+ "hidden_chart_oracle_utility": 0.2974560856819153,
838
+ "outcome_ptr": 0.0,
839
+ "predicted_margin": 0.4534903350362056,
840
+ "proposal_oracle_success": 0.0,
841
+ "proposal_oracle_utility": 0.060871422290802,
842
+ "selected_candidate_index": 0.0,
843
+ "selected_success": 0.0,
844
+ "selected_success_gain_over_base": 0.0,
845
+ "selected_utility": 0.0260902252048254,
846
+ "selected_utility_gain_over_base": -0.07842461206018925,
847
+ "selector_regret": 0.0347811970859766,
848
+ "success_selector_gap": 0.0,
849
+ "success_support_gap": 0.0,
850
+ "support_gap": 0.23658466339111328,
851
+ "tau": 0.3473188188503786
852
+ },
853
+ "1549324805/train1": {
854
+ "base_success": 0.0,
855
+ "base_utility": 0.10451483726501465,
856
+ "coverage": 0.0,
857
+ "execute_generated": 0.0,
858
+ "fallback_rate": 1.0,
859
+ "hidden_chart_oracle_success": 0.0,
860
+ "hidden_chart_oracle_utility": 0.2974560856819153,
861
+ "outcome_ptr": 1.0,
862
+ "predicted_margin": 0.1768234215359843,
863
+ "proposal_oracle_success": 0.0,
864
+ "proposal_oracle_utility": 0.19892151653766632,
865
+ "selected_candidate_index": 2.0,
866
+ "selected_success": 0.0,
867
+ "selected_success_gain_over_base": 0.0,
868
+ "selected_utility": 0.10451483726501465,
869
+ "selected_utility_gain_over_base": 0.0,
870
+ "selector_regret": 0.09440667927265167,
871
+ "success_selector_gap": 0.0,
872
+ "success_support_gap": 0.0,
873
+ "support_gap": 0.09853456914424896,
874
+ "tau": 0.3473188188503786
875
+ },
876
+ "1549324805/train2": {
877
+ "base_success": 0.0,
878
+ "base_utility": 0.10451483726501465,
879
+ "coverage": 1.0,
880
+ "execute_generated": 1.0,
881
+ "fallback_rate": 0.0,
882
+ "hidden_chart_oracle_success": 0.0,
883
+ "hidden_chart_oracle_utility": 0.2974560856819153,
884
+ "outcome_ptr": 1.0,
885
+ "predicted_margin": 0.5340001574841613,
886
+ "proposal_oracle_success": 0.0,
887
+ "proposal_oracle_utility": 0.11310184001922607,
888
+ "selected_candidate_index": 3.0,
889
+ "selected_success": 0.0,
890
+ "selected_success_gain_over_base": 0.0,
891
+ "selected_utility": 0.02059849165380001,
892
+ "selected_utility_gain_over_base": -0.08391634561121464,
893
+ "selector_regret": 0.09250334836542606,
894
+ "success_selector_gap": 0.0,
895
+ "success_support_gap": 0.0,
896
+ "support_gap": 0.1843542456626892,
897
+ "tau": 0.3473188188503786
898
+ },
899
+ "1578990937/train0": {
900
+ "base_success": 0.0,
901
+ "base_utility": 0.82413649559021,
902
+ "coverage": 0.0,
903
+ "execute_generated": 0.0,
904
+ "fallback_rate": 1.0,
905
+ "hidden_chart_oracle_success": 1.0,
906
+ "hidden_chart_oracle_utility": 2.0,
907
+ "outcome_ptr": 1.0,
908
+ "predicted_margin": 0.2255580985362039,
909
+ "proposal_oracle_success": 1.0,
910
+ "proposal_oracle_utility": 2.0,
911
+ "selected_candidate_index": 15.0,
912
+ "selected_success": 0.0,
913
+ "selected_success_gain_over_base": 0.0,
914
+ "selected_utility": 0.82413649559021,
915
+ "selected_utility_gain_over_base": 0.0,
916
+ "selector_regret": 1.17586350440979,
917
+ "success_selector_gap": 1.0,
918
+ "success_support_gap": 0.0,
919
+ "support_gap": 0.0,
920
+ "tau": 0.3473188188503786
921
+ },
922
+ "1578990937/train1": {
923
+ "base_success": 0.0,
924
+ "base_utility": 0.82413649559021,
925
+ "coverage": 0.0,
926
+ "execute_generated": 0.0,
927
+ "fallback_rate": 1.0,
928
+ "hidden_chart_oracle_success": 1.0,
929
+ "hidden_chart_oracle_utility": 2.0,
930
+ "outcome_ptr": 1.0,
931
+ "predicted_margin": 0.21684013213485268,
932
+ "proposal_oracle_success": 1.0,
933
+ "proposal_oracle_utility": 2.0,
934
+ "selected_candidate_index": 12.0,
935
+ "selected_success": 0.0,
936
+ "selected_success_gain_over_base": 0.0,
937
+ "selected_utility": 0.82413649559021,
938
+ "selected_utility_gain_over_base": 0.0,
939
+ "selector_regret": 1.17586350440979,
940
+ "success_selector_gap": 1.0,
941
+ "success_support_gap": 0.0,
942
+ "support_gap": 0.0,
943
+ "tau": 0.3473188188503786
944
+ },
945
+ "1578990937/train2": {
946
+ "base_success": 0.0,
947
+ "base_utility": 0.82413649559021,
948
+ "coverage": 0.0,
949
+ "execute_generated": 0.0,
950
+ "fallback_rate": 1.0,
951
+ "hidden_chart_oracle_success": 1.0,
952
+ "hidden_chart_oracle_utility": 2.0,
953
+ "outcome_ptr": 1.0,
954
+ "predicted_margin": 0.2250806530775492,
955
+ "proposal_oracle_success": 1.0,
956
+ "proposal_oracle_utility": 2.0,
957
+ "selected_candidate_index": 0.0,
958
+ "selected_success": 0.0,
959
+ "selected_success_gain_over_base": 0.0,
960
+ "selected_utility": 0.82413649559021,
961
+ "selected_utility_gain_over_base": 0.0,
962
+ "selector_regret": 1.17586350440979,
963
+ "success_selector_gap": 1.0,
964
+ "success_support_gap": 0.0,
965
+ "support_gap": 0.0,
966
+ "tau": 0.3473188188503786
967
+ },
968
+ "1747178030/train0": {
969
+ "base_success": 1.0,
970
+ "base_utility": 2.0,
971
+ "coverage": 0.0,
972
+ "execute_generated": 0.0,
973
+ "fallback_rate": 1.0,
974
+ "hidden_chart_oracle_success": 1.0,
975
+ "hidden_chart_oracle_utility": 2.0,
976
+ "outcome_ptr": 0.0,
977
+ "predicted_margin": 0.3352371313866231,
978
+ "proposal_oracle_success": 1.0,
979
+ "proposal_oracle_utility": 2.0,
980
+ "selected_candidate_index": 5.0,
981
+ "selected_success": 1.0,
982
+ "selected_success_gain_over_base": 0.0,
983
+ "selected_utility": 2.0,
984
+ "selected_utility_gain_over_base": 0.0,
985
+ "selector_regret": 0.0,
986
+ "success_selector_gap": 0.0,
987
+ "success_support_gap": 0.0,
988
+ "support_gap": 0.0,
989
+ "tau": 0.3473188188503786
990
+ },
991
+ "1747178030/train1": {
992
+ "base_success": 1.0,
993
+ "base_utility": 2.0,
994
+ "coverage": 0.0,
995
+ "execute_generated": 0.0,
996
+ "fallback_rate": 1.0,
997
+ "hidden_chart_oracle_success": 1.0,
998
+ "hidden_chart_oracle_utility": 2.0,
999
+ "outcome_ptr": 0.0,
1000
+ "predicted_margin": 0.33370414261502274,
1001
+ "proposal_oracle_success": 1.0,
1002
+ "proposal_oracle_utility": 2.0,
1003
+ "selected_candidate_index": 9.0,
1004
+ "selected_success": 1.0,
1005
+ "selected_success_gain_over_base": 0.0,
1006
+ "selected_utility": 2.0,
1007
+ "selected_utility_gain_over_base": 0.0,
1008
+ "selector_regret": 0.0,
1009
+ "success_selector_gap": 0.0,
1010
+ "success_support_gap": 0.0,
1011
+ "support_gap": 0.0,
1012
+ "tau": 0.3473188188503786
1013
+ },
1014
+ "1747178030/train2": {
1015
+ "base_success": 1.0,
1016
+ "base_utility": 2.0,
1017
+ "coverage": 0.0,
1018
+ "execute_generated": 0.0,
1019
+ "fallback_rate": 1.0,
1020
+ "hidden_chart_oracle_success": 1.0,
1021
+ "hidden_chart_oracle_utility": 2.0,
1022
+ "outcome_ptr": 0.0,
1023
+ "predicted_margin": 0.33681729037438607,
1024
+ "proposal_oracle_success": 1.0,
1025
+ "proposal_oracle_utility": 2.0,
1026
+ "selected_candidate_index": 0.0,
1027
+ "selected_success": 1.0,
1028
+ "selected_success_gain_over_base": 0.0,
1029
+ "selected_utility": 2.0,
1030
+ "selected_utility_gain_over_base": 0.0,
1031
+ "selector_regret": 0.0,
1032
+ "success_selector_gap": 0.0,
1033
+ "success_support_gap": 0.0,
1034
+ "support_gap": 0.0,
1035
+ "tau": 0.3473188188503786
1036
+ },
1037
+ "1790188787/train0": {
1038
+ "base_success": 0.0,
1039
+ "base_utility": 0.12476746737957001,
1040
+ "coverage": 1.0,
1041
+ "execute_generated": 1.0,
1042
+ "fallback_rate": 0.0,
1043
+ "hidden_chart_oracle_success": 1.0,
1044
+ "hidden_chart_oracle_utility": 2.0,
1045
+ "outcome_ptr": 1.0,
1046
+ "predicted_margin": 0.7256808008161852,
1047
+ "proposal_oracle_success": 1.0,
1048
+ "proposal_oracle_utility": 2.0,
1049
+ "selected_candidate_index": 9.0,
1050
+ "selected_success": 0.0,
1051
+ "selected_success_gain_over_base": 0.0,
1052
+ "selected_utility": 0.14965972304344177,
1053
+ "selected_utility_gain_over_base": 0.024892255663871765,
1054
+ "selector_regret": 1.8503402769565582,
1055
+ "success_selector_gap": 1.0,
1056
+ "success_support_gap": 0.0,
1057
+ "support_gap": 0.0,
1058
+ "tau": 0.3473188188503786
1059
+ },
1060
+ "1790188787/train1": {
1061
+ "base_success": 0.0,
1062
+ "base_utility": 0.12476746737957001,
1063
+ "coverage": 1.0,
1064
+ "execute_generated": 1.0,
1065
+ "fallback_rate": 0.0,
1066
+ "hidden_chart_oracle_success": 1.0,
1067
+ "hidden_chart_oracle_utility": 2.0,
1068
+ "outcome_ptr": 1.0,
1069
+ "predicted_margin": 0.7407334225672018,
1070
+ "proposal_oracle_success": 1.0,
1071
+ "proposal_oracle_utility": 2.0,
1072
+ "selected_candidate_index": 6.0,
1073
+ "selected_success": 0.0,
1074
+ "selected_success_gain_over_base": 0.0,
1075
+ "selected_utility": 0.13741976022720337,
1076
+ "selected_utility_gain_over_base": 0.012652292847633362,
1077
+ "selector_regret": 1.8625802397727966,
1078
+ "success_selector_gap": 1.0,
1079
+ "success_support_gap": 0.0,
1080
+ "support_gap": 0.0,
1081
+ "tau": 0.3473188188503786
1082
+ },
1083
+ "1790188787/train2": {
1084
+ "base_success": 0.0,
1085
+ "base_utility": 0.12476746737957001,
1086
+ "coverage": 1.0,
1087
+ "execute_generated": 1.0,
1088
+ "fallback_rate": 0.0,
1089
+ "hidden_chart_oracle_success": 1.0,
1090
+ "hidden_chart_oracle_utility": 2.0,
1091
+ "outcome_ptr": 1.0,
1092
+ "predicted_margin": 0.7188844098969164,
1093
+ "proposal_oracle_success": 1.0,
1094
+ "proposal_oracle_utility": 2.0,
1095
+ "selected_candidate_index": 0.0,
1096
+ "selected_success": 1.0,
1097
+ "selected_success_gain_over_base": 1.0,
1098
+ "selected_utility": 2.0,
1099
+ "selected_utility_gain_over_base": 1.87523253262043,
1100
+ "selector_regret": 0.0,
1101
+ "success_selector_gap": 0.0,
1102
+ "success_support_gap": 0.0,
1103
+ "support_gap": 0.0,
1104
+ "tau": 0.3473188188503786
1105
+ },
1106
+ "1822210471/train0": {
1107
+ "base_success": 0.0,
1108
+ "base_utility": 0.07191530615091324,
1109
+ "coverage": 0.0,
1110
+ "execute_generated": 0.0,
1111
+ "fallback_rate": 1.0,
1112
+ "hidden_chart_oracle_success": 1.0,
1113
+ "hidden_chart_oracle_utility": 2.0,
1114
+ "outcome_ptr": 1.0,
1115
+ "predicted_margin": 0.10685249697353241,
1116
+ "proposal_oracle_success": 1.0,
1117
+ "proposal_oracle_utility": 2.0,
1118
+ "selected_candidate_index": 8.0,
1119
+ "selected_success": 0.0,
1120
+ "selected_success_gain_over_base": 0.0,
1121
+ "selected_utility": 0.07191530615091324,
1122
+ "selected_utility_gain_over_base": 0.0,
1123
+ "selector_regret": 1.9280846938490868,
1124
+ "success_selector_gap": 1.0,
1125
+ "success_support_gap": 0.0,
1126
+ "support_gap": 0.0,
1127
+ "tau": 0.3473188188503786
1128
+ },
1129
+ "1822210471/train1": {
1130
+ "base_success": 0.0,
1131
+ "base_utility": 0.07191530615091324,
1132
+ "coverage": 0.0,
1133
+ "execute_generated": 0.0,
1134
+ "fallback_rate": 1.0,
1135
+ "hidden_chart_oracle_success": 1.0,
1136
+ "hidden_chart_oracle_utility": 2.0,
1137
+ "outcome_ptr": 1.0,
1138
+ "predicted_margin": 0.11932515699499706,
1139
+ "proposal_oracle_success": 1.0,
1140
+ "proposal_oracle_utility": 2.0,
1141
+ "selected_candidate_index": 5.0,
1142
+ "selected_success": 0.0,
1143
+ "selected_success_gain_over_base": 0.0,
1144
+ "selected_utility": 0.07191530615091324,
1145
+ "selected_utility_gain_over_base": 0.0,
1146
+ "selector_regret": 1.9280846938490868,
1147
+ "success_selector_gap": 1.0,
1148
+ "success_support_gap": 0.0,
1149
+ "support_gap": 0.0,
1150
+ "tau": 0.3473188188503786
1151
+ },
1152
+ "1822210471/train2": {
1153
+ "base_success": 0.0,
1154
+ "base_utility": 0.07191530615091324,
1155
+ "coverage": 0.0,
1156
+ "execute_generated": 0.0,
1157
+ "fallback_rate": 1.0,
1158
+ "hidden_chart_oracle_success": 1.0,
1159
+ "hidden_chart_oracle_utility": 2.0,
1160
+ "outcome_ptr": 1.0,
1161
+ "predicted_margin": 0.12054968766555163,
1162
+ "proposal_oracle_success": 1.0,
1163
+ "proposal_oracle_utility": 2.0,
1164
+ "selected_candidate_index": 0.0,
1165
+ "selected_success": 0.0,
1166
+ "selected_success_gain_over_base": 0.0,
1167
+ "selected_utility": 0.07191530615091324,
1168
+ "selected_utility_gain_over_base": 0.0,
1169
+ "selector_regret": 1.9280846938490868,
1170
+ "success_selector_gap": 1.0,
1171
+ "success_support_gap": 0.0,
1172
+ "support_gap": 0.0,
1173
+ "tau": 0.3473188188503786
1174
+ },
1175
+ "1845489013/train0": {
1176
+ "base_success": 1.0,
1177
+ "base_utility": 2.0,
1178
+ "coverage": 0.0,
1179
+ "execute_generated": 0.0,
1180
+ "fallback_rate": 1.0,
1181
+ "hidden_chart_oracle_success": 1.0,
1182
+ "hidden_chart_oracle_utility": 2.0,
1183
+ "outcome_ptr": 0.0,
1184
+ "predicted_margin": 0.22822097403519162,
1185
+ "proposal_oracle_success": 1.0,
1186
+ "proposal_oracle_utility": 2.0,
1187
+ "selected_candidate_index": 5.0,
1188
+ "selected_success": 1.0,
1189
+ "selected_success_gain_over_base": 0.0,
1190
+ "selected_utility": 2.0,
1191
+ "selected_utility_gain_over_base": 0.0,
1192
+ "selector_regret": 0.0,
1193
+ "success_selector_gap": 0.0,
1194
+ "success_support_gap": 0.0,
1195
+ "support_gap": 0.0,
1196
+ "tau": 0.3473188188503786
1197
+ },
1198
+ "1845489013/train1": {
1199
+ "base_success": 1.0,
1200
+ "base_utility": 2.0,
1201
+ "coverage": 0.0,
1202
+ "execute_generated": 0.0,
1203
+ "fallback_rate": 1.0,
1204
+ "hidden_chart_oracle_success": 1.0,
1205
+ "hidden_chart_oracle_utility": 2.0,
1206
+ "outcome_ptr": 0.0,
1207
+ "predicted_margin": 0.23290921447763308,
1208
+ "proposal_oracle_success": 1.0,
1209
+ "proposal_oracle_utility": 2.0,
1210
+ "selected_candidate_index": 7.0,
1211
+ "selected_success": 1.0,
1212
+ "selected_success_gain_over_base": 0.0,
1213
+ "selected_utility": 2.0,
1214
+ "selected_utility_gain_over_base": 0.0,
1215
+ "selector_regret": 0.0,
1216
+ "success_selector_gap": 0.0,
1217
+ "success_support_gap": 0.0,
1218
+ "support_gap": 0.0,
1219
+ "tau": 0.3473188188503786
1220
+ },
1221
+ "1845489013/train2": {
1222
+ "base_success": 1.0,
1223
+ "base_utility": 2.0,
1224
+ "coverage": 0.0,
1225
+ "execute_generated": 0.0,
1226
+ "fallback_rate": 1.0,
1227
+ "hidden_chart_oracle_success": 1.0,
1228
+ "hidden_chart_oracle_utility": 2.0,
1229
+ "outcome_ptr": 0.0,
1230
+ "predicted_margin": 0.23237107122905468,
1231
+ "proposal_oracle_success": 1.0,
1232
+ "proposal_oracle_utility": 2.0,
1233
+ "selected_candidate_index": 8.0,
1234
+ "selected_success": 1.0,
1235
+ "selected_success_gain_over_base": 0.0,
1236
+ "selected_utility": 2.0,
1237
+ "selected_utility_gain_over_base": 0.0,
1238
+ "selector_regret": 0.0,
1239
+ "success_selector_gap": 0.0,
1240
+ "success_support_gap": 0.0,
1241
+ "support_gap": 0.0,
1242
+ "tau": 0.3473188188503786
1243
+ },
1244
+ "1999022574/train0": {
1245
+ "base_success": 1.0,
1246
+ "base_utility": 2.0,
1247
+ "coverage": 0.0,
1248
+ "execute_generated": 0.0,
1249
+ "fallback_rate": 1.0,
1250
+ "hidden_chart_oracle_success": 1.0,
1251
+ "hidden_chart_oracle_utility": 2.0,
1252
+ "outcome_ptr": 0.0,
1253
+ "predicted_margin": 0.2830091920285253,
1254
+ "proposal_oracle_success": 1.0,
1255
+ "proposal_oracle_utility": 2.0,
1256
+ "selected_candidate_index": 11.0,
1257
+ "selected_success": 1.0,
1258
+ "selected_success_gain_over_base": 0.0,
1259
+ "selected_utility": 2.0,
1260
+ "selected_utility_gain_over_base": 0.0,
1261
+ "selector_regret": 0.0,
1262
+ "success_selector_gap": 0.0,
1263
+ "success_support_gap": 0.0,
1264
+ "support_gap": 0.0,
1265
+ "tau": 0.3473188188503786
1266
+ },
1267
+ "1999022574/train1": {
1268
+ "base_success": 1.0,
1269
+ "base_utility": 2.0,
1270
+ "coverage": 0.0,
1271
+ "execute_generated": 0.0,
1272
+ "fallback_rate": 1.0,
1273
+ "hidden_chart_oracle_success": 1.0,
1274
+ "hidden_chart_oracle_utility": 2.0,
1275
+ "outcome_ptr": 0.0,
1276
+ "predicted_margin": 0.2762933949412652,
1277
+ "proposal_oracle_success": 1.0,
1278
+ "proposal_oracle_utility": 2.0,
1279
+ "selected_candidate_index": 9.0,
1280
+ "selected_success": 1.0,
1281
+ "selected_success_gain_over_base": 0.0,
1282
+ "selected_utility": 2.0,
1283
+ "selected_utility_gain_over_base": 0.0,
1284
+ "selector_regret": 0.0,
1285
+ "success_selector_gap": 0.0,
1286
+ "success_support_gap": 0.0,
1287
+ "support_gap": 0.0,
1288
+ "tau": 0.3473188188503786
1289
+ },
1290
+ "1999022574/train2": {
1291
+ "base_success": 1.0,
1292
+ "base_utility": 2.0,
1293
+ "coverage": 0.0,
1294
+ "execute_generated": 0.0,
1295
+ "fallback_rate": 1.0,
1296
+ "hidden_chart_oracle_success": 1.0,
1297
+ "hidden_chart_oracle_utility": 2.0,
1298
+ "outcome_ptr": 0.0,
1299
+ "predicted_margin": 0.32995590919912554,
1300
+ "proposal_oracle_success": 1.0,
1301
+ "proposal_oracle_utility": 2.0,
1302
+ "selected_candidate_index": 6.0,
1303
+ "selected_success": 1.0,
1304
+ "selected_success_gain_over_base": 0.0,
1305
+ "selected_utility": 2.0,
1306
+ "selected_utility_gain_over_base": 0.0,
1307
+ "selector_regret": 0.0,
1308
+ "success_selector_gap": 0.0,
1309
+ "success_support_gap": 0.0,
1310
+ "support_gap": 0.0,
1311
+ "tau": 0.3473188188503786
1312
+ },
1313
+ "2012211287/train0": {
1314
+ "base_success": 0.0,
1315
+ "base_utility": 0.7236053347587585,
1316
+ "coverage": 0.0,
1317
+ "execute_generated": 0.0,
1318
+ "fallback_rate": 1.0,
1319
+ "hidden_chart_oracle_success": 1.0,
1320
+ "hidden_chart_oracle_utility": 2.0,
1321
+ "outcome_ptr": 1.0,
1322
+ "predicted_margin": 0.08991720318616597,
1323
+ "proposal_oracle_success": 1.0,
1324
+ "proposal_oracle_utility": 2.0,
1325
+ "selected_candidate_index": 0.0,
1326
+ "selected_success": 0.0,
1327
+ "selected_success_gain_over_base": 0.0,
1328
+ "selected_utility": 0.7236053347587585,
1329
+ "selected_utility_gain_over_base": 0.0,
1330
+ "selector_regret": 1.2763946652412415,
1331
+ "success_selector_gap": 1.0,
1332
+ "success_support_gap": 0.0,
1333
+ "support_gap": 0.0,
1334
+ "tau": 0.3473188188503786
1335
+ },
1336
+ "2012211287/train1": {
1337
+ "base_success": 0.0,
1338
+ "base_utility": 0.7236053347587585,
1339
+ "coverage": 0.0,
1340
+ "execute_generated": 0.0,
1341
+ "fallback_rate": 1.0,
1342
+ "hidden_chart_oracle_success": 1.0,
1343
+ "hidden_chart_oracle_utility": 2.0,
1344
+ "outcome_ptr": 1.0,
1345
+ "predicted_margin": 0.1018745030033766,
1346
+ "proposal_oracle_success": 1.0,
1347
+ "proposal_oracle_utility": 2.0,
1348
+ "selected_candidate_index": 5.0,
1349
+ "selected_success": 0.0,
1350
+ "selected_success_gain_over_base": 0.0,
1351
+ "selected_utility": 0.7236053347587585,
1352
+ "selected_utility_gain_over_base": 0.0,
1353
+ "selector_regret": 1.2763946652412415,
1354
+ "success_selector_gap": 1.0,
1355
+ "success_support_gap": 0.0,
1356
+ "support_gap": 0.0,
1357
+ "tau": 0.3473188188503786
1358
+ },
1359
+ "2012211287/train2": {
1360
+ "base_success": 0.0,
1361
+ "base_utility": 0.7236053347587585,
1362
+ "coverage": 0.0,
1363
+ "execute_generated": 0.0,
1364
+ "fallback_rate": 1.0,
1365
+ "hidden_chart_oracle_success": 1.0,
1366
+ "hidden_chart_oracle_utility": 2.0,
1367
+ "outcome_ptr": 1.0,
1368
+ "predicted_margin": 0.09743460718887578,
1369
+ "proposal_oracle_success": 1.0,
1370
+ "proposal_oracle_utility": 2.0,
1371
+ "selected_candidate_index": 1.0,
1372
+ "selected_success": 0.0,
1373
+ "selected_success_gain_over_base": 0.0,
1374
+ "selected_utility": 0.7236053347587585,
1375
+ "selected_utility_gain_over_base": 0.0,
1376
+ "selector_regret": 1.2763946652412415,
1377
+ "success_selector_gap": 1.0,
1378
+ "success_support_gap": 0.0,
1379
+ "support_gap": 0.0,
1380
+ "tau": 0.3473188188503786
1381
+ },
1382
+ "2038594191/train0": {
1383
+ "base_success": 1.0,
1384
+ "base_utility": 2.0,
1385
+ "coverage": 1.0,
1386
+ "execute_generated": 1.0,
1387
+ "fallback_rate": 0.0,
1388
+ "hidden_chart_oracle_success": 1.0,
1389
+ "hidden_chart_oracle_utility": 2.0,
1390
+ "outcome_ptr": 0.0,
1391
+ "predicted_margin": 0.41980384640853585,
1392
+ "proposal_oracle_success": 1.0,
1393
+ "proposal_oracle_utility": 2.0,
1394
+ "selected_candidate_index": 3.0,
1395
+ "selected_success": 0.0,
1396
+ "selected_success_gain_over_base": -1.0,
1397
+ "selected_utility": 0.05387979745864868,
1398
+ "selected_utility_gain_over_base": -1.9461202025413513,
1399
+ "selector_regret": 1.9461202025413513,
1400
+ "success_selector_gap": 1.0,
1401
+ "success_support_gap": 0.0,
1402
+ "support_gap": 0.0,
1403
+ "tau": 0.3473188188503786
1404
+ },
1405
+ "2038594191/train1": {
1406
+ "base_success": 1.0,
1407
+ "base_utility": 2.0,
1408
+ "coverage": 1.0,
1409
+ "execute_generated": 1.0,
1410
+ "fallback_rate": 0.0,
1411
+ "hidden_chart_oracle_success": 1.0,
1412
+ "hidden_chart_oracle_utility": 2.0,
1413
+ "outcome_ptr": 0.0,
1414
+ "predicted_margin": 0.39227093434151067,
1415
+ "proposal_oracle_success": 1.0,
1416
+ "proposal_oracle_utility": 2.0,
1417
+ "selected_candidate_index": 0.0,
1418
+ "selected_success": 0.0,
1419
+ "selected_success_gain_over_base": -1.0,
1420
+ "selected_utility": 0.023384133353829384,
1421
+ "selected_utility_gain_over_base": -1.9766158666461706,
1422
+ "selector_regret": 1.9766158666461706,
1423
+ "success_selector_gap": 1.0,
1424
+ "success_support_gap": 0.0,
1425
+ "support_gap": 0.0,
1426
+ "tau": 0.3473188188503786
1427
+ },
1428
+ "2038594191/train2": {
1429
+ "base_success": 1.0,
1430
+ "base_utility": 2.0,
1431
+ "coverage": 1.0,
1432
+ "execute_generated": 1.0,
1433
+ "fallback_rate": 0.0,
1434
+ "hidden_chart_oracle_success": 1.0,
1435
+ "hidden_chart_oracle_utility": 2.0,
1436
+ "outcome_ptr": 0.0,
1437
+ "predicted_margin": 0.41980384640853585,
1438
+ "proposal_oracle_success": 1.0,
1439
+ "proposal_oracle_utility": 2.0,
1440
+ "selected_candidate_index": 1.0,
1441
+ "selected_success": 1.0,
1442
+ "selected_success_gain_over_base": 0.0,
1443
+ "selected_utility": 2.0,
1444
+ "selected_utility_gain_over_base": 0.0,
1445
+ "selector_regret": 0.0,
1446
+ "success_selector_gap": 0.0,
1447
+ "success_support_gap": 0.0,
1448
+ "support_gap": 0.0,
1449
+ "tau": 0.3473188188503786
1450
+ },
1451
+ "2054899126/train0": {
1452
+ "base_success": 1.0,
1453
+ "base_utility": 2.0,
1454
+ "coverage": 1.0,
1455
+ "execute_generated": 1.0,
1456
+ "fallback_rate": 0.0,
1457
+ "hidden_chart_oracle_success": 1.0,
1458
+ "hidden_chart_oracle_utility": 2.0,
1459
+ "outcome_ptr": 0.0,
1460
+ "predicted_margin": 0.48682363081170954,
1461
+ "proposal_oracle_success": 0.5,
1462
+ "proposal_oracle_utility": 1.077283151447773,
1463
+ "selected_candidate_index": 0.0,
1464
+ "selected_success": 0.5,
1465
+ "selected_success_gain_over_base": -0.5,
1466
+ "selected_utility": 1.052448581904173,
1467
+ "selected_utility_gain_over_base": -0.9475514180958271,
1468
+ "selector_regret": 0.024834569543600082,
1469
+ "success_selector_gap": 0.0,
1470
+ "success_support_gap": 0.5,
1471
+ "support_gap": 0.922716848552227,
1472
+ "tau": 0.3473188188503786
1473
+ },
1474
+ "2054899126/train1": {
1475
+ "base_success": 1.0,
1476
+ "base_utility": 2.0,
1477
+ "coverage": 1.0,
1478
+ "execute_generated": 1.0,
1479
+ "fallback_rate": 0.0,
1480
+ "hidden_chart_oracle_success": 1.0,
1481
+ "hidden_chart_oracle_utility": 2.0,
1482
+ "outcome_ptr": 0.0,
1483
+ "predicted_margin": 0.4867067691213679,
1484
+ "proposal_oracle_success": 1.0,
1485
+ "proposal_oracle_utility": 2.0,
1486
+ "selected_candidate_index": 0.0,
1487
+ "selected_success": 0.5,
1488
+ "selected_success_gain_over_base": -0.5,
1489
+ "selected_utility": 1.0664558708667755,
1490
+ "selected_utility_gain_over_base": -0.9335441291332245,
1491
+ "selector_regret": 0.9335441291332245,
1492
+ "success_selector_gap": 0.5,
1493
+ "success_support_gap": 0.0,
1494
+ "support_gap": 0.0,
1495
+ "tau": 0.3473188188503786
1496
+ },
1497
+ "2054899126/train2": {
1498
+ "base_success": 1.0,
1499
+ "base_utility": 2.0,
1500
+ "coverage": 0.5,
1501
+ "execute_generated": 0.5,
1502
+ "fallback_rate": 0.5,
1503
+ "hidden_chart_oracle_success": 1.0,
1504
+ "hidden_chart_oracle_utility": 2.0,
1505
+ "outcome_ptr": 0.0,
1506
+ "predicted_margin": 0.4238219648796263,
1507
+ "proposal_oracle_success": 1.0,
1508
+ "proposal_oracle_utility": 2.0,
1509
+ "selected_candidate_index": 0.5,
1510
+ "selected_success": 1.0,
1511
+ "selected_success_gain_over_base": 0.0,
1512
+ "selected_utility": 2.0,
1513
+ "selected_utility_gain_over_base": 0.0,
1514
+ "selector_regret": 0.0,
1515
+ "success_selector_gap": 0.0,
1516
+ "success_support_gap": 0.0,
1517
+ "support_gap": 0.0,
1518
+ "tau": 0.3473188188503786
1519
+ },
1520
+ "2061635888/train0": {
1521
+ "base_success": 0.0,
1522
+ "base_utility": 0.05225050449371338,
1523
+ "coverage": 0.0,
1524
+ "execute_generated": 0.0,
1525
+ "fallback_rate": 1.0,
1526
+ "hidden_chart_oracle_success": 1.0,
1527
+ "hidden_chart_oracle_utility": 2.0,
1528
+ "outcome_ptr": 1.0,
1529
+ "predicted_margin": 0.16138880459399857,
1530
+ "proposal_oracle_success": 0.0,
1531
+ "proposal_oracle_utility": 0.09799930453300476,
1532
+ "selected_candidate_index": 5.0,
1533
+ "selected_success": 0.0,
1534
+ "selected_success_gain_over_base": 0.0,
1535
+ "selected_utility": 0.05225050449371338,
1536
+ "selected_utility_gain_over_base": 0.0,
1537
+ "selector_regret": 0.04574880003929138,
1538
+ "success_selector_gap": 0.0,
1539
+ "success_support_gap": 1.0,
1540
+ "support_gap": 1.9020006954669952,
1541
+ "tau": 0.3473188188503786
1542
+ },
1543
+ "2061635888/train1": {
1544
+ "base_success": 0.0,
1545
+ "base_utility": 0.05225050449371338,
1546
+ "coverage": 0.0,
1547
+ "execute_generated": 0.0,
1548
+ "fallback_rate": 1.0,
1549
+ "hidden_chart_oracle_success": 1.0,
1550
+ "hidden_chart_oracle_utility": 2.0,
1551
+ "outcome_ptr": 1.0,
1552
+ "predicted_margin": 0.12788381852052721,
1553
+ "proposal_oracle_success": 0.0,
1554
+ "proposal_oracle_utility": 0.07580336928367615,
1555
+ "selected_candidate_index": 6.0,
1556
+ "selected_success": 0.0,
1557
+ "selected_success_gain_over_base": 0.0,
1558
+ "selected_utility": 0.05225050449371338,
1559
+ "selected_utility_gain_over_base": 0.0,
1560
+ "selector_regret": 0.02355286478996277,
1561
+ "success_selector_gap": 0.0,
1562
+ "success_support_gap": 1.0,
1563
+ "support_gap": 1.9241966307163239,
1564
+ "tau": 0.3473188188503786
1565
+ },
1566
+ "2061635888/train2": {
1567
+ "base_success": 0.0,
1568
+ "base_utility": 0.05225050449371338,
1569
+ "coverage": 0.0,
1570
+ "execute_generated": 0.0,
1571
+ "fallback_rate": 1.0,
1572
+ "hidden_chart_oracle_success": 1.0,
1573
+ "hidden_chart_oracle_utility": 2.0,
1574
+ "outcome_ptr": 1.0,
1575
+ "predicted_margin": 0.13169501516378973,
1576
+ "proposal_oracle_success": 0.0,
1577
+ "proposal_oracle_utility": 0.11105132102966309,
1578
+ "selected_candidate_index": 4.0,
1579
+ "selected_success": 0.0,
1580
+ "selected_success_gain_over_base": 0.0,
1581
+ "selected_utility": 0.05225050449371338,
1582
+ "selected_utility_gain_over_base": 0.0,
1583
+ "selector_regret": 0.05880081653594971,
1584
+ "success_selector_gap": 0.0,
1585
+ "success_support_gap": 1.0,
1586
+ "support_gap": 1.888948678970337,
1587
+ "tau": 0.3473188188503786
1588
+ },
1589
+ "211245842/train0": {
1590
+ "base_success": 0.0,
1591
+ "base_utility": 0.19090841710567474,
1592
+ "coverage": 1.0,
1593
+ "execute_generated": 1.0,
1594
+ "fallback_rate": 0.0,
1595
+ "hidden_chart_oracle_success": 1.0,
1596
+ "hidden_chart_oracle_utility": 2.0,
1597
+ "outcome_ptr": 1.0,
1598
+ "predicted_margin": 0.36446911237871227,
1599
+ "proposal_oracle_success": 1.0,
1600
+ "proposal_oracle_utility": 2.0,
1601
+ "selected_candidate_index": 7.0,
1602
+ "selected_success": 0.0,
1603
+ "selected_success_gain_over_base": 0.0,
1604
+ "selected_utility": 0.11833827197551727,
1605
+ "selected_utility_gain_over_base": -0.07257014513015747,
1606
+ "selector_regret": 1.8816617280244827,
1607
+ "success_selector_gap": 1.0,
1608
+ "success_support_gap": 0.0,
1609
+ "support_gap": 0.0,
1610
+ "tau": 0.3473188188503786
1611
+ },
1612
+ "211245842/train1": {
1613
+ "base_success": 0.0,
1614
+ "base_utility": 0.19090841710567474,
1615
+ "coverage": 1.0,
1616
+ "execute_generated": 1.0,
1617
+ "fallback_rate": 0.0,
1618
+ "hidden_chart_oracle_success": 1.0,
1619
+ "hidden_chart_oracle_utility": 2.0,
1620
+ "outcome_ptr": 1.0,
1621
+ "predicted_margin": 0.3818735579959133,
1622
+ "proposal_oracle_success": 0.0,
1623
+ "proposal_oracle_utility": 0.21295717358589172,
1624
+ "selected_candidate_index": 13.0,
1625
+ "selected_success": 0.0,
1626
+ "selected_success_gain_over_base": 0.0,
1627
+ "selected_utility": 0.17156320810317993,
1628
+ "selected_utility_gain_over_base": -0.019345209002494812,
1629
+ "selector_regret": 0.04139396548271179,
1630
+ "success_selector_gap": 0.0,
1631
+ "success_support_gap": 1.0,
1632
+ "support_gap": 1.7870428264141083,
1633
+ "tau": 0.3473188188503786
1634
+ },
1635
+ "211245842/train2": {
1636
+ "base_success": 0.0,
1637
+ "base_utility": 0.19090841710567474,
1638
+ "coverage": 1.0,
1639
+ "execute_generated": 1.0,
1640
+ "fallback_rate": 0.0,
1641
+ "hidden_chart_oracle_success": 1.0,
1642
+ "hidden_chart_oracle_utility": 2.0,
1643
+ "outcome_ptr": 1.0,
1644
+ "predicted_margin": 0.3818735579959133,
1645
+ "proposal_oracle_success": 0.0,
1646
+ "proposal_oracle_utility": 0.23388276994228363,
1647
+ "selected_candidate_index": 0.0,
1648
+ "selected_success": 0.0,
1649
+ "selected_success_gain_over_base": 0.0,
1650
+ "selected_utility": 0.17691531777381897,
1651
+ "selected_utility_gain_over_base": -0.013993099331855774,
1652
+ "selector_regret": 0.05696745216846466,
1653
+ "success_selector_gap": 0.0,
1654
+ "success_support_gap": 1.0,
1655
+ "support_gap": 1.7661172300577164,
1656
+ "tau": 0.3473188188503786
1657
+ },
1658
+ "278008524/train0": {
1659
+ "base_success": 0.0,
1660
+ "base_utility": 0.042734989896416664,
1661
+ "coverage": 1.0,
1662
+ "execute_generated": 1.0,
1663
+ "fallback_rate": 0.0,
1664
+ "hidden_chart_oracle_success": 0.5,
1665
+ "hidden_chart_oracle_utility": 1.1139086410403252,
1666
+ "outcome_ptr": 1.0,
1667
+ "predicted_margin": 0.4154810514015457,
1668
+ "proposal_oracle_success": 0.5,
1669
+ "proposal_oracle_utility": 1.0733563229441643,
1670
+ "selected_candidate_index": 3.0,
1671
+ "selected_success": 0.5,
1672
+ "selected_success_gain_over_base": 0.5,
1673
+ "selected_utility": 1.0485825426876545,
1674
+ "selected_utility_gain_over_base": 1.0058475527912378,
1675
+ "selector_regret": 0.02477378025650978,
1676
+ "success_selector_gap": 0.0,
1677
+ "success_support_gap": 0.0,
1678
+ "support_gap": 0.04055231809616089,
1679
+ "tau": 0.3473188188503786
1680
+ },
1681
+ "278008524/train1": {
1682
+ "base_success": 0.0,
1683
+ "base_utility": 0.042734989896416664,
1684
+ "coverage": 1.0,
1685
+ "execute_generated": 1.0,
1686
+ "fallback_rate": 0.0,
1687
+ "hidden_chart_oracle_success": 0.5,
1688
+ "hidden_chart_oracle_utility": 1.1139086410403252,
1689
+ "outcome_ptr": 1.0,
1690
+ "predicted_margin": 0.431062905506948,
1691
+ "proposal_oracle_success": 0.5,
1692
+ "proposal_oracle_utility": 1.0376745946705341,
1693
+ "selected_candidate_index": 1.0,
1694
+ "selected_success": 0.0,
1695
+ "selected_success_gain_over_base": 0.0,
1696
+ "selected_utility": 0.1001356840133667,
1697
+ "selected_utility_gain_over_base": 0.057400694116950035,
1698
+ "selector_regret": 0.9375389106571674,
1699
+ "success_selector_gap": 0.5,
1700
+ "success_support_gap": 0.0,
1701
+ "support_gap": 0.07623404636979103,
1702
+ "tau": 0.3473188188503786
1703
+ },
1704
+ "278008524/train2": {
1705
+ "base_success": 0.0,
1706
+ "base_utility": 0.042734989896416664,
1707
+ "coverage": 0.5,
1708
+ "execute_generated": 0.5,
1709
+ "fallback_rate": 0.5,
1710
+ "hidden_chart_oracle_success": 0.5,
1711
+ "hidden_chart_oracle_utility": 1.1139086410403252,
1712
+ "outcome_ptr": 1.0,
1713
+ "predicted_margin": 0.39125281310385696,
1714
+ "proposal_oracle_success": 0.5,
1715
+ "proposal_oracle_utility": 1.1359489113092422,
1716
+ "selected_candidate_index": 11.5,
1717
+ "selected_success": 0.0,
1718
+ "selected_success_gain_over_base": 0.0,
1719
+ "selected_utility": 0.052897630259394646,
1720
+ "selected_utility_gain_over_base": 0.010162640362977982,
1721
+ "selector_regret": 1.0830512810498476,
1722
+ "success_selector_gap": 0.5,
1723
+ "success_support_gap": 0.0,
1724
+ "support_gap": 0.0,
1725
+ "tau": 0.3473188188503786
1726
+ },
1727
+ "307068879/train0": {
1728
+ "base_success": 0.0,
1729
+ "base_utility": 0.6000905632972717,
1730
+ "coverage": 0.0,
1731
+ "execute_generated": 0.0,
1732
+ "fallback_rate": 1.0,
1733
+ "hidden_chart_oracle_success": 1.0,
1734
+ "hidden_chart_oracle_utility": 2.0,
1735
+ "outcome_ptr": 0.0,
1736
+ "predicted_margin": 0.2797080480367002,
1737
+ "proposal_oracle_success": 0.0,
1738
+ "proposal_oracle_utility": 0.2115575522184372,
1739
+ "selected_candidate_index": 2.0,
1740
+ "selected_success": 0.0,
1741
+ "selected_success_gain_over_base": 0.0,
1742
+ "selected_utility": 0.6000905632972717,
1743
+ "selected_utility_gain_over_base": 0.0,
1744
+ "selector_regret": 0.0,
1745
+ "success_selector_gap": 0.0,
1746
+ "success_support_gap": 1.0,
1747
+ "support_gap": 1.7884424477815628,
1748
+ "tau": 0.3473188188503786
1749
+ },
1750
+ "307068879/train1": {
1751
+ "base_success": 0.0,
1752
+ "base_utility": 0.6000905632972717,
1753
+ "coverage": 0.0,
1754
+ "execute_generated": 0.0,
1755
+ "fallback_rate": 1.0,
1756
+ "hidden_chart_oracle_success": 1.0,
1757
+ "hidden_chart_oracle_utility": 2.0,
1758
+ "outcome_ptr": 0.0,
1759
+ "predicted_margin": 0.24778606153341362,
1760
+ "proposal_oracle_success": 0.0,
1761
+ "proposal_oracle_utility": 0.5876073241233826,
1762
+ "selected_candidate_index": 12.0,
1763
+ "selected_success": 0.0,
1764
+ "selected_success_gain_over_base": 0.0,
1765
+ "selected_utility": 0.6000905632972717,
1766
+ "selected_utility_gain_over_base": 0.0,
1767
+ "selector_regret": 0.0,
1768
+ "success_selector_gap": 0.0,
1769
+ "success_support_gap": 1.0,
1770
+ "support_gap": 1.4123926758766174,
1771
+ "tau": 0.3473188188503786
1772
+ },
1773
+ "307068879/train2": {
1774
+ "base_success": 0.0,
1775
+ "base_utility": 0.6000905632972717,
1776
+ "coverage": 0.0,
1777
+ "execute_generated": 0.0,
1778
+ "fallback_rate": 1.0,
1779
+ "hidden_chart_oracle_success": 1.0,
1780
+ "hidden_chart_oracle_utility": 2.0,
1781
+ "outcome_ptr": 0.0,
1782
+ "predicted_margin": 0.29487688188584205,
1783
+ "proposal_oracle_success": 0.0,
1784
+ "proposal_oracle_utility": 0.2302134484052658,
1785
+ "selected_candidate_index": 0.0,
1786
+ "selected_success": 0.0,
1787
+ "selected_success_gain_over_base": 0.0,
1788
+ "selected_utility": 0.6000905632972717,
1789
+ "selected_utility_gain_over_base": 0.0,
1790
+ "selector_regret": 0.0,
1791
+ "success_selector_gap": 0.0,
1792
+ "success_support_gap": 1.0,
1793
+ "support_gap": 1.7697865515947342,
1794
+ "tau": 0.3473188188503786
1795
+ },
1796
+ "412102563/train0": {
1797
+ "base_success": 1.0,
1798
+ "base_utility": 2.0,
1799
+ "coverage": 0.0,
1800
+ "execute_generated": 0.0,
1801
+ "fallback_rate": 1.0,
1802
+ "hidden_chart_oracle_success": 1.0,
1803
+ "hidden_chart_oracle_utility": 2.0,
1804
+ "outcome_ptr": 0.0,
1805
+ "predicted_margin": 0.20585303592937382,
1806
+ "proposal_oracle_success": 1.0,
1807
+ "proposal_oracle_utility": 2.0,
1808
+ "selected_candidate_index": 6.0,
1809
+ "selected_success": 1.0,
1810
+ "selected_success_gain_over_base": 0.0,
1811
+ "selected_utility": 2.0,
1812
+ "selected_utility_gain_over_base": 0.0,
1813
+ "selector_regret": 0.0,
1814
+ "success_selector_gap": 0.0,
1815
+ "success_support_gap": 0.0,
1816
+ "support_gap": 0.0,
1817
+ "tau": 0.3473188188503786
1818
+ },
1819
+ "412102563/train1": {
1820
+ "base_success": 1.0,
1821
+ "base_utility": 2.0,
1822
+ "coverage": 0.0,
1823
+ "execute_generated": 0.0,
1824
+ "fallback_rate": 1.0,
1825
+ "hidden_chart_oracle_success": 1.0,
1826
+ "hidden_chart_oracle_utility": 2.0,
1827
+ "outcome_ptr": 0.0,
1828
+ "predicted_margin": 0.20493799872831076,
1829
+ "proposal_oracle_success": 1.0,
1830
+ "proposal_oracle_utility": 2.0,
1831
+ "selected_candidate_index": 10.0,
1832
+ "selected_success": 1.0,
1833
+ "selected_success_gain_over_base": 0.0,
1834
+ "selected_utility": 2.0,
1835
+ "selected_utility_gain_over_base": 0.0,
1836
+ "selector_regret": 0.0,
1837
+ "success_selector_gap": 0.0,
1838
+ "success_support_gap": 0.0,
1839
+ "support_gap": 0.0,
1840
+ "tau": 0.3473188188503786
1841
+ },
1842
+ "412102563/train2": {
1843
+ "base_success": 1.0,
1844
+ "base_utility": 2.0,
1845
+ "coverage": 0.0,
1846
+ "execute_generated": 0.0,
1847
+ "fallback_rate": 1.0,
1848
+ "hidden_chart_oracle_success": 1.0,
1849
+ "hidden_chart_oracle_utility": 2.0,
1850
+ "outcome_ptr": 0.0,
1851
+ "predicted_margin": 0.18125860724557305,
1852
+ "proposal_oracle_success": 1.0,
1853
+ "proposal_oracle_utility": 2.0,
1854
+ "selected_candidate_index": 3.0,
1855
+ "selected_success": 1.0,
1856
+ "selected_success_gain_over_base": 0.0,
1857
+ "selected_utility": 2.0,
1858
+ "selected_utility_gain_over_base": 0.0,
1859
+ "selector_regret": 0.0,
1860
+ "success_selector_gap": 0.0,
1861
+ "success_support_gap": 0.0,
1862
+ "support_gap": 0.0,
1863
+ "tau": 0.3473188188503786
1864
+ },
1865
+ "440577711/train0": {
1866
+ "base_success": 0.0,
1867
+ "base_utility": 0.036152780055999756,
1868
+ "coverage": 0.0,
1869
+ "execute_generated": 0.0,
1870
+ "fallback_rate": 1.0,
1871
+ "hidden_chart_oracle_success": 1.0,
1872
+ "hidden_chart_oracle_utility": 2.0,
1873
+ "outcome_ptr": 1.0,
1874
+ "predicted_margin": 0.1872847414980603,
1875
+ "proposal_oracle_success": 1.0,
1876
+ "proposal_oracle_utility": 2.0,
1877
+ "selected_candidate_index": 3.0,
1878
+ "selected_success": 0.0,
1879
+ "selected_success_gain_over_base": 0.0,
1880
+ "selected_utility": 0.036152780055999756,
1881
+ "selected_utility_gain_over_base": 0.0,
1882
+ "selector_regret": 1.9638472199440002,
1883
+ "success_selector_gap": 1.0,
1884
+ "success_support_gap": 0.0,
1885
+ "support_gap": 0.0,
1886
+ "tau": 0.3473188188503786
1887
+ },
1888
+ "440577711/train1": {
1889
+ "base_success": 0.0,
1890
+ "base_utility": 0.036152780055999756,
1891
+ "coverage": 0.0,
1892
+ "execute_generated": 0.0,
1893
+ "fallback_rate": 1.0,
1894
+ "hidden_chart_oracle_success": 1.0,
1895
+ "hidden_chart_oracle_utility": 2.0,
1896
+ "outcome_ptr": 1.0,
1897
+ "predicted_margin": 0.1872847414980603,
1898
+ "proposal_oracle_success": 1.0,
1899
+ "proposal_oracle_utility": 2.0,
1900
+ "selected_candidate_index": 5.0,
1901
+ "selected_success": 0.0,
1902
+ "selected_success_gain_over_base": 0.0,
1903
+ "selected_utility": 0.036152780055999756,
1904
+ "selected_utility_gain_over_base": 0.0,
1905
+ "selector_regret": 1.9638472199440002,
1906
+ "success_selector_gap": 1.0,
1907
+ "success_support_gap": 0.0,
1908
+ "support_gap": 0.0,
1909
+ "tau": 0.3473188188503786
1910
+ },
1911
+ "440577711/train2": {
1912
+ "base_success": 0.0,
1913
+ "base_utility": 0.036152780055999756,
1914
+ "coverage": 0.0,
1915
+ "execute_generated": 0.0,
1916
+ "fallback_rate": 1.0,
1917
+ "hidden_chart_oracle_success": 1.0,
1918
+ "hidden_chart_oracle_utility": 2.0,
1919
+ "outcome_ptr": 1.0,
1920
+ "predicted_margin": 0.1872847414980603,
1921
+ "proposal_oracle_success": 1.0,
1922
+ "proposal_oracle_utility": 2.0,
1923
+ "selected_candidate_index": 5.0,
1924
+ "selected_success": 0.0,
1925
+ "selected_success_gain_over_base": 0.0,
1926
+ "selected_utility": 0.036152780055999756,
1927
+ "selected_utility_gain_over_base": 0.0,
1928
+ "selector_regret": 1.9638472199440002,
1929
+ "success_selector_gap": 1.0,
1930
+ "success_support_gap": 0.0,
1931
+ "support_gap": 0.0,
1932
+ "tau": 0.3473188188503786
1933
+ },
1934
+ "441046872/train0": {
1935
+ "base_success": 0.0,
1936
+ "base_utility": 0.6999551653862,
1937
+ "coverage": 0.0,
1938
+ "execute_generated": 0.0,
1939
+ "fallback_rate": 1.0,
1940
+ "hidden_chart_oracle_success": 1.0,
1941
+ "hidden_chart_oracle_utility": 2.0,
1942
+ "outcome_ptr": 1.0,
1943
+ "predicted_margin": 0.20172868695503152,
1944
+ "proposal_oracle_success": 1.0,
1945
+ "proposal_oracle_utility": 2.0,
1946
+ "selected_candidate_index": 10.0,
1947
+ "selected_success": 0.0,
1948
+ "selected_success_gain_over_base": 0.0,
1949
+ "selected_utility": 0.6999551653862,
1950
+ "selected_utility_gain_over_base": 0.0,
1951
+ "selector_regret": 1.3000448346138,
1952
+ "success_selector_gap": 1.0,
1953
+ "success_support_gap": 0.0,
1954
+ "support_gap": 0.0,
1955
+ "tau": 0.3473188188503786
1956
+ },
1957
+ "441046872/train1": {
1958
+ "base_success": 0.0,
1959
+ "base_utility": 0.6999551653862,
1960
+ "coverage": 0.0,
1961
+ "execute_generated": 0.0,
1962
+ "fallback_rate": 1.0,
1963
+ "hidden_chart_oracle_success": 1.0,
1964
+ "hidden_chart_oracle_utility": 2.0,
1965
+ "outcome_ptr": 1.0,
1966
+ "predicted_margin": 0.24443771108596596,
1967
+ "proposal_oracle_success": 0.0,
1968
+ "proposal_oracle_utility": 0.7233588695526123,
1969
+ "selected_candidate_index": 3.0,
1970
+ "selected_success": 0.0,
1971
+ "selected_success_gain_over_base": 0.0,
1972
+ "selected_utility": 0.6999551653862,
1973
+ "selected_utility_gain_over_base": 0.0,
1974
+ "selector_regret": 0.023403704166412354,
1975
+ "success_selector_gap": 0.0,
1976
+ "success_support_gap": 1.0,
1977
+ "support_gap": 1.2766411304473877,
1978
+ "tau": 0.3473188188503786
1979
+ },
1980
+ "441046872/train2": {
1981
+ "base_success": 0.0,
1982
+ "base_utility": 0.6999551653862,
1983
+ "coverage": 0.0,
1984
+ "execute_generated": 0.0,
1985
+ "fallback_rate": 1.0,
1986
+ "hidden_chart_oracle_success": 1.0,
1987
+ "hidden_chart_oracle_utility": 2.0,
1988
+ "outcome_ptr": 1.0,
1989
+ "predicted_margin": 0.19710526958727811,
1990
+ "proposal_oracle_success": 0.0,
1991
+ "proposal_oracle_utility": 0.7191348671913147,
1992
+ "selected_candidate_index": 0.0,
1993
+ "selected_success": 0.0,
1994
+ "selected_success_gain_over_base": 0.0,
1995
+ "selected_utility": 0.6999551653862,
1996
+ "selected_utility_gain_over_base": 0.0,
1997
+ "selector_regret": 0.019179701805114746,
1998
+ "success_selector_gap": 0.0,
1999
+ "success_support_gap": 1.0,
2000
+ "support_gap": 1.2808651328086853,
2001
+ "tau": 0.3473188188503786
2002
+ },
2003
+ "509430392/train0": {
2004
+ "base_success": 0.0,
2005
+ "base_utility": 2.2351741790771484e-07,
2006
+ "coverage": 1.0,
2007
+ "execute_generated": 1.0,
2008
+ "fallback_rate": 0.0,
2009
+ "hidden_chart_oracle_success": 1.0,
2010
+ "hidden_chart_oracle_utility": 2.0,
2011
+ "outcome_ptr": 1.0,
2012
+ "predicted_margin": 0.39582323274190295,
2013
+ "proposal_oracle_success": 1.0,
2014
+ "proposal_oracle_utility": 2.0,
2015
+ "selected_candidate_index": 14.0,
2016
+ "selected_success": 0.0,
2017
+ "selected_success_gain_over_base": 0.0,
2018
+ "selected_utility": 0.12097799777984619,
2019
+ "selected_utility_gain_over_base": 0.12097777426242828,
2020
+ "selector_regret": 1.8790220022201538,
2021
+ "success_selector_gap": 1.0,
2022
+ "success_support_gap": 0.0,
2023
+ "support_gap": 0.0,
2024
+ "tau": 0.3473188188503786
2025
+ },
2026
+ "509430392/train1": {
2027
+ "base_success": 0.0,
2028
+ "base_utility": 2.2351741790771484e-07,
2029
+ "coverage": 1.0,
2030
+ "execute_generated": 1.0,
2031
+ "fallback_rate": 0.0,
2032
+ "hidden_chart_oracle_success": 1.0,
2033
+ "hidden_chart_oracle_utility": 2.0,
2034
+ "outcome_ptr": 1.0,
2035
+ "predicted_margin": 0.3836105423742797,
2036
+ "proposal_oracle_success": 1.0,
2037
+ "proposal_oracle_utility": 2.0,
2038
+ "selected_candidate_index": 2.0,
2039
+ "selected_success": 1.0,
2040
+ "selected_success_gain_over_base": 1.0,
2041
+ "selected_utility": 2.0,
2042
+ "selected_utility_gain_over_base": 1.999999776482582,
2043
+ "selector_regret": 0.0,
2044
+ "success_selector_gap": 0.0,
2045
+ "success_support_gap": 0.0,
2046
+ "support_gap": 0.0,
2047
+ "tau": 0.3473188188503786
2048
+ },
2049
+ "509430392/train2": {
2050
+ "base_success": 0.0,
2051
+ "base_utility": 2.2351741790771484e-07,
2052
+ "coverage": 1.0,
2053
+ "execute_generated": 1.0,
2054
+ "fallback_rate": 0.0,
2055
+ "hidden_chart_oracle_success": 1.0,
2056
+ "hidden_chart_oracle_utility": 2.0,
2057
+ "outcome_ptr": 1.0,
2058
+ "predicted_margin": 0.3982631871077599,
2059
+ "proposal_oracle_success": 0.0,
2060
+ "proposal_oracle_utility": 5.424022674560547e-06,
2061
+ "selected_candidate_index": 1.0,
2062
+ "selected_success": 0.0,
2063
+ "selected_success_gain_over_base": 0.0,
2064
+ "selected_utility": 2.682209014892578e-07,
2065
+ "selected_utility_gain_over_base": 4.470348358154297e-08,
2066
+ "selector_regret": 5.155801773071289e-06,
2067
+ "success_selector_gap": 0.0,
2068
+ "success_support_gap": 1.0,
2069
+ "support_gap": 1.9999945759773254,
2070
+ "tau": 0.3473188188503786
2071
+ },
2072
+ "585904130/train0": {
2073
+ "base_success": 0.0,
2074
+ "base_utility": 0.18171554803848267,
2075
+ "coverage": 0.0,
2076
+ "execute_generated": 0.0,
2077
+ "fallback_rate": 1.0,
2078
+ "hidden_chart_oracle_success": 1.0,
2079
+ "hidden_chart_oracle_utility": 2.0,
2080
+ "outcome_ptr": 1.0,
2081
+ "predicted_margin": 0.26430287849596273,
2082
+ "proposal_oracle_success": 1.0,
2083
+ "proposal_oracle_utility": 2.0,
2084
+ "selected_candidate_index": 14.0,
2085
+ "selected_success": 0.0,
2086
+ "selected_success_gain_over_base": 0.0,
2087
+ "selected_utility": 0.18171554803848267,
2088
+ "selected_utility_gain_over_base": 0.0,
2089
+ "selector_regret": 1.8182844519615173,
2090
+ "success_selector_gap": 1.0,
2091
+ "success_support_gap": 0.0,
2092
+ "support_gap": 0.0,
2093
+ "tau": 0.3473188188503786
2094
+ },
2095
+ "585904130/train1": {
2096
+ "base_success": 0.0,
2097
+ "base_utility": 0.18171554803848267,
2098
+ "coverage": 0.0,
2099
+ "execute_generated": 0.0,
2100
+ "fallback_rate": 1.0,
2101
+ "hidden_chart_oracle_success": 1.0,
2102
+ "hidden_chart_oracle_utility": 2.0,
2103
+ "outcome_ptr": 1.0,
2104
+ "predicted_margin": 0.23332367249842212,
2105
+ "proposal_oracle_success": 1.0,
2106
+ "proposal_oracle_utility": 2.0,
2107
+ "selected_candidate_index": 7.0,
2108
+ "selected_success": 0.0,
2109
+ "selected_success_gain_over_base": 0.0,
2110
+ "selected_utility": 0.18171554803848267,
2111
+ "selected_utility_gain_over_base": 0.0,
2112
+ "selector_regret": 1.8182844519615173,
2113
+ "success_selector_gap": 1.0,
2114
+ "success_support_gap": 0.0,
2115
+ "support_gap": 0.0,
2116
+ "tau": 0.3473188188503786
2117
+ },
2118
+ "585904130/train2": {
2119
+ "base_success": 0.0,
2120
+ "base_utility": 0.18171554803848267,
2121
+ "coverage": 0.0,
2122
+ "execute_generated": 0.0,
2123
+ "fallback_rate": 1.0,
2124
+ "hidden_chart_oracle_success": 1.0,
2125
+ "hidden_chart_oracle_utility": 2.0,
2126
+ "outcome_ptr": 1.0,
2127
+ "predicted_margin": 0.2531734715082984,
2128
+ "proposal_oracle_success": 1.0,
2129
+ "proposal_oracle_utility": 2.0,
2130
+ "selected_candidate_index": 4.0,
2131
+ "selected_success": 0.0,
2132
+ "selected_success_gain_over_base": 0.0,
2133
+ "selected_utility": 0.18171554803848267,
2134
+ "selected_utility_gain_over_base": 0.0,
2135
+ "selector_regret": 1.8182844519615173,
2136
+ "success_selector_gap": 1.0,
2137
+ "success_support_gap": 0.0,
2138
+ "support_gap": 0.0,
2139
+ "tau": 0.3473188188503786
2140
+ },
2141
+ "608016290/train0": {
2142
+ "base_success": 0.0,
2143
+ "base_utility": 0.725701630115509,
2144
+ "coverage": 0.0,
2145
+ "execute_generated": 0.0,
2146
+ "fallback_rate": 1.0,
2147
+ "hidden_chart_oracle_success": 1.0,
2148
+ "hidden_chart_oracle_utility": 2.0,
2149
+ "outcome_ptr": 1.0,
2150
+ "predicted_margin": 0.18560038716365618,
2151
+ "proposal_oracle_success": 1.0,
2152
+ "proposal_oracle_utility": 2.0,
2153
+ "selected_candidate_index": 3.0,
2154
+ "selected_success": 0.0,
2155
+ "selected_success_gain_over_base": 0.0,
2156
+ "selected_utility": 0.725701630115509,
2157
+ "selected_utility_gain_over_base": 0.0,
2158
+ "selector_regret": 1.274298369884491,
2159
+ "success_selector_gap": 1.0,
2160
+ "success_support_gap": 0.0,
2161
+ "support_gap": 0.0,
2162
+ "tau": 0.3473188188503786
2163
+ },
2164
+ "608016290/train1": {
2165
+ "base_success": 0.0,
2166
+ "base_utility": 0.725701630115509,
2167
+ "coverage": 0.0,
2168
+ "execute_generated": 0.0,
2169
+ "fallback_rate": 1.0,
2170
+ "hidden_chart_oracle_success": 1.0,
2171
+ "hidden_chart_oracle_utility": 2.0,
2172
+ "outcome_ptr": 1.0,
2173
+ "predicted_margin": 0.2379598925580104,
2174
+ "proposal_oracle_success": 1.0,
2175
+ "proposal_oracle_utility": 2.0,
2176
+ "selected_candidate_index": 8.0,
2177
+ "selected_success": 0.0,
2178
+ "selected_success_gain_over_base": 0.0,
2179
+ "selected_utility": 0.725701630115509,
2180
+ "selected_utility_gain_over_base": 0.0,
2181
+ "selector_regret": 1.274298369884491,
2182
+ "success_selector_gap": 1.0,
2183
+ "success_support_gap": 0.0,
2184
+ "support_gap": 0.0,
2185
+ "tau": 0.3473188188503786
2186
+ },
2187
+ "608016290/train2": {
2188
+ "base_success": 0.0,
2189
+ "base_utility": 0.725701630115509,
2190
+ "coverage": 0.0,
2191
+ "execute_generated": 0.0,
2192
+ "fallback_rate": 1.0,
2193
+ "hidden_chart_oracle_success": 1.0,
2194
+ "hidden_chart_oracle_utility": 2.0,
2195
+ "outcome_ptr": 0.0,
2196
+ "predicted_margin": 0.1808325968136315,
2197
+ "proposal_oracle_success": 0.0,
2198
+ "proposal_oracle_utility": 0.7117276191711426,
2199
+ "selected_candidate_index": 15.0,
2200
+ "selected_success": 0.0,
2201
+ "selected_success_gain_over_base": 0.0,
2202
+ "selected_utility": 0.725701630115509,
2203
+ "selected_utility_gain_over_base": 0.0,
2204
+ "selector_regret": 0.0,
2205
+ "success_selector_gap": 0.0,
2206
+ "success_support_gap": 1.0,
2207
+ "support_gap": 1.2882723808288574,
2208
+ "tau": 0.3473188188503786
2209
+ },
2210
+ "626486813/train0": {
2211
+ "base_success": 0.0,
2212
+ "base_utility": 0.00509834272088483,
2213
+ "coverage": 0.5,
2214
+ "execute_generated": 0.5,
2215
+ "fallback_rate": 0.5,
2216
+ "hidden_chart_oracle_success": 0.0,
2217
+ "hidden_chart_oracle_utility": 0.16558238118886948,
2218
+ "outcome_ptr": 1.0,
2219
+ "predicted_margin": 0.41319658430741574,
2220
+ "proposal_oracle_success": 0.0,
2221
+ "proposal_oracle_utility": 0.04469547560438514,
2222
+ "selected_candidate_index": 0.0,
2223
+ "selected_success": 0.0,
2224
+ "selected_success_gain_over_base": 0.0,
2225
+ "selected_utility": 0.005386819248087704,
2226
+ "selected_utility_gain_over_base": 0.0002884765272028744,
2227
+ "selector_regret": 0.03930865635629743,
2228
+ "success_selector_gap": 0.0,
2229
+ "success_support_gap": 0.0,
2230
+ "support_gap": 0.12161866808310151,
2231
+ "tau": 0.3473188188503786
2232
+ },
2233
+ "626486813/train1": {
2234
+ "base_success": 0.0,
2235
+ "base_utility": 0.00509834272088483,
2236
+ "coverage": 0.5,
2237
+ "execute_generated": 0.5,
2238
+ "fallback_rate": 0.5,
2239
+ "hidden_chart_oracle_success": 0.0,
2240
+ "hidden_chart_oracle_utility": 0.16558238118886948,
2241
+ "outcome_ptr": 1.0,
2242
+ "predicted_margin": 0.46817656253554407,
2243
+ "proposal_oracle_success": 0.0,
2244
+ "proposal_oracle_utility": 0.035827874671667814,
2245
+ "selected_candidate_index": 0.5,
2246
+ "selected_success": 0.0,
2247
+ "selected_success_gain_over_base": 0.0,
2248
+ "selected_utility": 0.00787139916792512,
2249
+ "selected_utility_gain_over_base": 0.0027730564470402896,
2250
+ "selector_regret": 0.027956475503742695,
2251
+ "success_selector_gap": 0.0,
2252
+ "success_support_gap": 0.0,
2253
+ "support_gap": 0.12975450651720166,
2254
+ "tau": 0.3473188188503786
2255
+ },
2256
+ "626486813/train2": {
2257
+ "base_success": 0.0,
2258
+ "base_utility": 0.00509834272088483,
2259
+ "coverage": 0.5,
2260
+ "execute_generated": 0.5,
2261
+ "fallback_rate": 0.5,
2262
+ "hidden_chart_oracle_success": 0.0,
2263
+ "hidden_chart_oracle_utility": 0.16558238118886948,
2264
+ "outcome_ptr": 1.0,
2265
+ "predicted_margin": 0.3841254242297905,
2266
+ "proposal_oracle_success": 0.0,
2267
+ "proposal_oracle_utility": 0.04489925503730774,
2268
+ "selected_candidate_index": 1.0,
2269
+ "selected_success": 0.0,
2270
+ "selected_success_gain_over_base": 0.0,
2271
+ "selected_utility": 0.015375842805951834,
2272
+ "selected_utility_gain_over_base": 0.010277500085067004,
2273
+ "selector_regret": 0.029523412231355906,
2274
+ "success_selector_gap": 0.0,
2275
+ "success_support_gap": 0.0,
2276
+ "support_gap": 0.12068312615156174,
2277
+ "tau": 0.3473188188503786
2278
+ },
2279
+ "636680137/train0": {
2280
+ "base_success": 0.0,
2281
+ "base_utility": 0.08313504606485367,
2282
+ "coverage": 1.0,
2283
+ "execute_generated": 1.0,
2284
+ "fallback_rate": 0.0,
2285
+ "hidden_chart_oracle_success": 0.0,
2286
+ "hidden_chart_oracle_utility": 0.17056432366371155,
2287
+ "outcome_ptr": 1.0,
2288
+ "predicted_margin": 0.5371194860392053,
2289
+ "proposal_oracle_success": 0.0,
2290
+ "proposal_oracle_utility": 0.08826130628585815,
2291
+ "selected_candidate_index": 0.0,
2292
+ "selected_success": 0.0,
2293
+ "selected_success_gain_over_base": 0.0,
2294
+ "selected_utility": 0.056092143058776855,
2295
+ "selected_utility_gain_over_base": -0.027042903006076813,
2296
+ "selector_regret": 0.0321691632270813,
2297
+ "success_selector_gap": 0.0,
2298
+ "success_support_gap": 0.0,
2299
+ "support_gap": 0.0823030173778534,
2300
+ "tau": 0.3473188188503786
2301
+ },
2302
+ "636680137/train1": {
2303
+ "base_success": 0.0,
2304
+ "base_utility": 0.08313504606485367,
2305
+ "coverage": 0.0,
2306
+ "execute_generated": 0.0,
2307
+ "fallback_rate": 1.0,
2308
+ "hidden_chart_oracle_success": 0.0,
2309
+ "hidden_chart_oracle_utility": 0.17056432366371155,
2310
+ "outcome_ptr": 0.0,
2311
+ "predicted_margin": 0.2496823681236075,
2312
+ "proposal_oracle_success": 0.0,
2313
+ "proposal_oracle_utility": 0.06741118431091309,
2314
+ "selected_candidate_index": 0.0,
2315
+ "selected_success": 0.0,
2316
+ "selected_success_gain_over_base": 0.0,
2317
+ "selected_utility": 0.08313504606485367,
2318
+ "selected_utility_gain_over_base": 0.0,
2319
+ "selector_regret": 0.0,
2320
+ "success_selector_gap": 0.0,
2321
+ "success_support_gap": 0.0,
2322
+ "support_gap": 0.10315313935279846,
2323
+ "tau": 0.3473188188503786
2324
+ },
2325
+ "636680137/train2": {
2326
+ "base_success": 0.0,
2327
+ "base_utility": 0.08313504606485367,
2328
+ "coverage": 0.0,
2329
+ "execute_generated": 0.0,
2330
+ "fallback_rate": 1.0,
2331
+ "hidden_chart_oracle_success": 0.0,
2332
+ "hidden_chart_oracle_utility": 0.17056432366371155,
2333
+ "outcome_ptr": 0.0,
2334
+ "predicted_margin": 0.2912910141265051,
2335
+ "proposal_oracle_success": 0.0,
2336
+ "proposal_oracle_utility": 0.0756516233086586,
2337
+ "selected_candidate_index": 1.0,
2338
+ "selected_success": 0.0,
2339
+ "selected_success_gain_over_base": 0.0,
2340
+ "selected_utility": 0.08313504606485367,
2341
+ "selected_utility_gain_over_base": 0.0,
2342
+ "selector_regret": 0.0,
2343
+ "success_selector_gap": 0.0,
2344
+ "success_support_gap": 0.0,
2345
+ "support_gap": 0.09491270035505295,
2346
+ "tau": 0.3473188188503786
2347
+ },
2348
+ "655884605/train0": {
2349
+ "base_success": 0.0,
2350
+ "base_utility": 0.16300667822360992,
2351
+ "coverage": 0.0,
2352
+ "execute_generated": 0.0,
2353
+ "fallback_rate": 1.0,
2354
+ "hidden_chart_oracle_success": 1.0,
2355
+ "hidden_chart_oracle_utility": 2.0,
2356
+ "outcome_ptr": 0.0,
2357
+ "predicted_margin": 0.30368062376907057,
2358
+ "proposal_oracle_success": 0.0,
2359
+ "proposal_oracle_utility": 0.13732868432998657,
2360
+ "selected_candidate_index": 1.0,
2361
+ "selected_success": 0.0,
2362
+ "selected_success_gain_over_base": 0.0,
2363
+ "selected_utility": 0.16300667822360992,
2364
+ "selected_utility_gain_over_base": 0.0,
2365
+ "selector_regret": 0.0,
2366
+ "success_selector_gap": 0.0,
2367
+ "success_support_gap": 1.0,
2368
+ "support_gap": 1.8626713156700134,
2369
+ "tau": 0.3473188188503786
2370
+ },
2371
+ "655884605/train1": {
2372
+ "base_success": 0.0,
2373
+ "base_utility": 0.16300667822360992,
2374
+ "coverage": 0.0,
2375
+ "execute_generated": 0.0,
2376
+ "fallback_rate": 1.0,
2377
+ "hidden_chart_oracle_success": 1.0,
2378
+ "hidden_chart_oracle_utility": 2.0,
2379
+ "outcome_ptr": 1.0,
2380
+ "predicted_margin": 0.2713166038796752,
2381
+ "proposal_oracle_success": 0.0,
2382
+ "proposal_oracle_utility": 0.8128758072853088,
2383
+ "selected_candidate_index": 2.0,
2384
+ "selected_success": 0.0,
2385
+ "selected_success_gain_over_base": 0.0,
2386
+ "selected_utility": 0.16300667822360992,
2387
+ "selected_utility_gain_over_base": 0.0,
2388
+ "selector_regret": 0.6498691290616989,
2389
+ "success_selector_gap": 0.0,
2390
+ "success_support_gap": 1.0,
2391
+ "support_gap": 1.1871241927146912,
2392
+ "tau": 0.3473188188503786
2393
+ },
2394
+ "655884605/train2": {
2395
+ "base_success": 0.0,
2396
+ "base_utility": 0.16300667822360992,
2397
+ "coverage": 0.0,
2398
+ "execute_generated": 0.0,
2399
+ "fallback_rate": 1.0,
2400
+ "hidden_chart_oracle_success": 1.0,
2401
+ "hidden_chart_oracle_utility": 2.0,
2402
+ "outcome_ptr": 1.0,
2403
+ "predicted_margin": 0.19620741382881715,
2404
+ "proposal_oracle_success": 0.0,
2405
+ "proposal_oracle_utility": 0.21447330713272095,
2406
+ "selected_candidate_index": 8.0,
2407
+ "selected_success": 0.0,
2408
+ "selected_success_gain_over_base": 0.0,
2409
+ "selected_utility": 0.16300667822360992,
2410
+ "selected_utility_gain_over_base": 0.0,
2411
+ "selector_regret": 0.05146662890911102,
2412
+ "success_selector_gap": 0.0,
2413
+ "success_support_gap": 1.0,
2414
+ "support_gap": 1.785526692867279,
2415
+ "tau": 0.3473188188503786
2416
+ },
2417
+ "672328262/train0": {
2418
+ "base_success": 1.0,
2419
+ "base_utility": 2.0,
2420
+ "coverage": 0.0,
2421
+ "execute_generated": 0.0,
2422
+ "fallback_rate": 1.0,
2423
+ "hidden_chart_oracle_success": 1.0,
2424
+ "hidden_chart_oracle_utility": 2.0,
2425
+ "outcome_ptr": 0.0,
2426
+ "predicted_margin": 0.042802515015988585,
2427
+ "proposal_oracle_success": 1.0,
2428
+ "proposal_oracle_utility": 2.0,
2429
+ "selected_candidate_index": 8.0,
2430
+ "selected_success": 1.0,
2431
+ "selected_success_gain_over_base": 0.0,
2432
+ "selected_utility": 2.0,
2433
+ "selected_utility_gain_over_base": 0.0,
2434
+ "selector_regret": 0.0,
2435
+ "success_selector_gap": 0.0,
2436
+ "success_support_gap": 0.0,
2437
+ "support_gap": 0.0,
2438
+ "tau": 0.3473188188503786
2439
+ },
2440
+ "672328262/train1": {
2441
+ "base_success": 1.0,
2442
+ "base_utility": 2.0,
2443
+ "coverage": 0.0,
2444
+ "execute_generated": 0.0,
2445
+ "fallback_rate": 1.0,
2446
+ "hidden_chart_oracle_success": 1.0,
2447
+ "hidden_chart_oracle_utility": 2.0,
2448
+ "outcome_ptr": 0.0,
2449
+ "predicted_margin": 0.04466194101303914,
2450
+ "proposal_oracle_success": 1.0,
2451
+ "proposal_oracle_utility": 2.0,
2452
+ "selected_candidate_index": 1.0,
2453
+ "selected_success": 1.0,
2454
+ "selected_success_gain_over_base": 0.0,
2455
+ "selected_utility": 2.0,
2456
+ "selected_utility_gain_over_base": 0.0,
2457
+ "selector_regret": 0.0,
2458
+ "success_selector_gap": 0.0,
2459
+ "success_support_gap": 0.0,
2460
+ "support_gap": 0.0,
2461
+ "tau": 0.3473188188503786
2462
+ },
2463
+ "672328262/train2": {
2464
+ "base_success": 1.0,
2465
+ "base_utility": 2.0,
2466
+ "coverage": 0.0,
2467
+ "execute_generated": 0.0,
2468
+ "fallback_rate": 1.0,
2469
+ "hidden_chart_oracle_success": 1.0,
2470
+ "hidden_chart_oracle_utility": 2.0,
2471
+ "outcome_ptr": 0.0,
2472
+ "predicted_margin": 0.048270193237023314,
2473
+ "proposal_oracle_success": 1.0,
2474
+ "proposal_oracle_utility": 2.0,
2475
+ "selected_candidate_index": 2.0,
2476
+ "selected_success": 1.0,
2477
+ "selected_success_gain_over_base": 0.0,
2478
+ "selected_utility": 2.0,
2479
+ "selected_utility_gain_over_base": 0.0,
2480
+ "selector_regret": 0.0,
2481
+ "success_selector_gap": 0.0,
2482
+ "success_support_gap": 0.0,
2483
+ "support_gap": 0.0,
2484
+ "tau": 0.3473188188503786
2485
+ },
2486
+ "701667486/train0": {
2487
+ "base_success": 1.0,
2488
+ "base_utility": 2.0,
2489
+ "coverage": 1.0,
2490
+ "execute_generated": 1.0,
2491
+ "fallback_rate": 0.0,
2492
+ "hidden_chart_oracle_success": 1.0,
2493
+ "hidden_chart_oracle_utility": 2.0,
2494
+ "outcome_ptr": 0.0,
2495
+ "predicted_margin": 0.3589247640152091,
2496
+ "proposal_oracle_success": 1.0,
2497
+ "proposal_oracle_utility": 2.0,
2498
+ "selected_candidate_index": 2.0,
2499
+ "selected_success": 1.0,
2500
+ "selected_success_gain_over_base": 0.0,
2501
+ "selected_utility": 2.0,
2502
+ "selected_utility_gain_over_base": 0.0,
2503
+ "selector_regret": 0.0,
2504
+ "success_selector_gap": 0.0,
2505
+ "success_support_gap": 0.0,
2506
+ "support_gap": 0.0,
2507
+ "tau": 0.3473188188503786
2508
+ },
2509
+ "701667486/train1": {
2510
+ "base_success": 1.0,
2511
+ "base_utility": 2.0,
2512
+ "coverage": 1.0,
2513
+ "execute_generated": 1.0,
2514
+ "fallback_rate": 0.0,
2515
+ "hidden_chart_oracle_success": 1.0,
2516
+ "hidden_chart_oracle_utility": 2.0,
2517
+ "outcome_ptr": 0.0,
2518
+ "predicted_margin": 0.38798745623031156,
2519
+ "proposal_oracle_success": 1.0,
2520
+ "proposal_oracle_utility": 2.0,
2521
+ "selected_candidate_index": 10.0,
2522
+ "selected_success": 1.0,
2523
+ "selected_success_gain_over_base": 0.0,
2524
+ "selected_utility": 2.0,
2525
+ "selected_utility_gain_over_base": 0.0,
2526
+ "selector_regret": 0.0,
2527
+ "success_selector_gap": 0.0,
2528
+ "success_support_gap": 0.0,
2529
+ "support_gap": 0.0,
2530
+ "tau": 0.3473188188503786
2531
+ },
2532
+ "701667486/train2": {
2533
+ "base_success": 1.0,
2534
+ "base_utility": 2.0,
2535
+ "coverage": 1.0,
2536
+ "execute_generated": 1.0,
2537
+ "fallback_rate": 0.0,
2538
+ "hidden_chart_oracle_success": 1.0,
2539
+ "hidden_chart_oracle_utility": 2.0,
2540
+ "outcome_ptr": 0.0,
2541
+ "predicted_margin": 0.4021201653187239,
2542
+ "proposal_oracle_success": 1.0,
2543
+ "proposal_oracle_utility": 2.0,
2544
+ "selected_candidate_index": 8.0,
2545
+ "selected_success": 1.0,
2546
+ "selected_success_gain_over_base": 0.0,
2547
+ "selected_utility": 2.0,
2548
+ "selected_utility_gain_over_base": 0.0,
2549
+ "selector_regret": 0.0,
2550
+ "success_selector_gap": 0.0,
2551
+ "success_support_gap": 0.0,
2552
+ "support_gap": 0.0,
2553
+ "tau": 0.3473188188503786
2554
+ },
2555
+ "747260437/train0": {
2556
+ "base_success": 1.0,
2557
+ "base_utility": 2.0,
2558
+ "coverage": 0.0,
2559
+ "execute_generated": 0.0,
2560
+ "fallback_rate": 1.0,
2561
+ "hidden_chart_oracle_success": 0.0,
2562
+ "hidden_chart_oracle_utility": 0.2786632478237152,
2563
+ "outcome_ptr": 0.0,
2564
+ "predicted_margin": 0.28297507531186483,
2565
+ "proposal_oracle_success": 0.0,
2566
+ "proposal_oracle_utility": 0.10839563608169556,
2567
+ "selected_candidate_index": 0.0,
2568
+ "selected_success": 1.0,
2569
+ "selected_success_gain_over_base": 0.0,
2570
+ "selected_utility": 2.0,
2571
+ "selected_utility_gain_over_base": 0.0,
2572
+ "selector_regret": 0.0,
2573
+ "success_selector_gap": 0.0,
2574
+ "success_support_gap": 0.0,
2575
+ "support_gap": 0.17026761174201965,
2576
+ "tau": 0.3473188188503786
2577
+ },
2578
+ "747260437/train1": {
2579
+ "base_success": 1.0,
2580
+ "base_utility": 2.0,
2581
+ "coverage": 0.0,
2582
+ "execute_generated": 0.0,
2583
+ "fallback_rate": 1.0,
2584
+ "hidden_chart_oracle_success": 0.0,
2585
+ "hidden_chart_oracle_utility": 0.2786632478237152,
2586
+ "outcome_ptr": 0.0,
2587
+ "predicted_margin": 0.2525195047503382,
2588
+ "proposal_oracle_success": 0.0,
2589
+ "proposal_oracle_utility": 0.06328952312469482,
2590
+ "selected_candidate_index": 8.0,
2591
+ "selected_success": 1.0,
2592
+ "selected_success_gain_over_base": 0.0,
2593
+ "selected_utility": 2.0,
2594
+ "selected_utility_gain_over_base": 0.0,
2595
+ "selector_regret": 0.0,
2596
+ "success_selector_gap": 0.0,
2597
+ "success_support_gap": 0.0,
2598
+ "support_gap": 0.21537372469902039,
2599
+ "tau": 0.3473188188503786
2600
+ },
2601
+ "747260437/train2": {
2602
+ "base_success": 1.0,
2603
+ "base_utility": 2.0,
2604
+ "coverage": 1.0,
2605
+ "execute_generated": 1.0,
2606
+ "fallback_rate": 0.0,
2607
+ "hidden_chart_oracle_success": 0.0,
2608
+ "hidden_chart_oracle_utility": 0.2786632478237152,
2609
+ "outcome_ptr": 0.0,
2610
+ "predicted_margin": 0.5259115214300562,
2611
+ "proposal_oracle_success": 0.0,
2612
+ "proposal_oracle_utility": 0.12510187923908234,
2613
+ "selected_candidate_index": 1.0,
2614
+ "selected_success": 0.0,
2615
+ "selected_success_gain_over_base": -1.0,
2616
+ "selected_utility": 0.06632735580205917,
2617
+ "selected_utility_gain_over_base": -1.9336726441979408,
2618
+ "selector_regret": 0.05877452343702316,
2619
+ "success_selector_gap": 0.0,
2620
+ "success_support_gap": 0.0,
2621
+ "support_gap": 0.15356136858463287,
2622
+ "tau": 0.3473188188503786
2623
+ },
2624
+ "750448296/train0": {
2625
+ "base_success": 0.0,
2626
+ "base_utility": 0.11759314934412639,
2627
+ "coverage": 0.3333333333333333,
2628
+ "execute_generated": 0.3333333333333333,
2629
+ "fallback_rate": 0.6666666666666666,
2630
+ "hidden_chart_oracle_success": 0.6666666666666666,
2631
+ "hidden_chart_oracle_utility": 1.4079964955647786,
2632
+ "outcome_ptr": 0.6666666666666666,
2633
+ "predicted_margin": 0.4190984327494606,
2634
+ "proposal_oracle_success": 0.3333333333333333,
2635
+ "proposal_oracle_utility": 0.7526603279014429,
2636
+ "selected_candidate_index": 4.0,
2637
+ "selected_success": 0.0,
2638
+ "selected_success_gain_over_base": 0.0,
2639
+ "selected_utility": 0.09317883538703124,
2640
+ "selected_utility_gain_over_base": -0.024414313957095146,
2641
+ "selector_regret": 0.6594814925144116,
2642
+ "success_selector_gap": 0.3333333333333333,
2643
+ "success_support_gap": 0.3333333333333333,
2644
+ "support_gap": 0.6553361676633358,
2645
+ "tau": 0.3473188188503786
2646
+ },
2647
+ "750448296/train1": {
2648
+ "base_success": 0.0,
2649
+ "base_utility": 0.11759314934412639,
2650
+ "coverage": 0.3333333333333333,
2651
+ "execute_generated": 0.3333333333333333,
2652
+ "fallback_rate": 0.6666666666666666,
2653
+ "hidden_chart_oracle_success": 0.6666666666666666,
2654
+ "hidden_chart_oracle_utility": 1.4079964955647786,
2655
+ "outcome_ptr": 1.0,
2656
+ "predicted_margin": 0.33489946089892014,
2657
+ "proposal_oracle_success": 0.6666666666666666,
2658
+ "proposal_oracle_utility": 1.3705032741030057,
2659
+ "selected_candidate_index": 3.0,
2660
+ "selected_success": 0.0,
2661
+ "selected_success_gain_over_base": 0.0,
2662
+ "selected_utility": 0.11618362615505855,
2663
+ "selected_utility_gain_over_base": -0.0014095231890678406,
2664
+ "selector_regret": 1.2543196479479473,
2665
+ "success_selector_gap": 0.6666666666666666,
2666
+ "success_support_gap": 0.0,
2667
+ "support_gap": 0.03749322146177292,
2668
+ "tau": 0.3473188188503786
2669
+ },
2670
+ "750448296/train2": {
2671
+ "base_success": 0.0,
2672
+ "base_utility": 0.11759314934412639,
2673
+ "coverage": 0.3333333333333333,
2674
+ "execute_generated": 0.3333333333333333,
2675
+ "fallback_rate": 0.6666666666666666,
2676
+ "hidden_chart_oracle_success": 0.6666666666666666,
2677
+ "hidden_chart_oracle_utility": 1.4079964955647786,
2678
+ "outcome_ptr": 0.6666666666666666,
2679
+ "predicted_margin": 0.31678157342373975,
2680
+ "proposal_oracle_success": 0.6666666666666666,
2681
+ "proposal_oracle_utility": 1.3504486680030823,
2682
+ "selected_candidate_index": 2.0,
2683
+ "selected_success": 0.0,
2684
+ "selected_success_gain_over_base": 0.0,
2685
+ "selected_utility": 0.1037093202273051,
2686
+ "selected_utility_gain_over_base": -0.013883829116821289,
2687
+ "selector_regret": 1.246739347775777,
2688
+ "success_selector_gap": 0.6666666666666666,
2689
+ "success_support_gap": 0.0,
2690
+ "support_gap": 0.05754782756169637,
2691
+ "tau": 0.3473188188503786
2692
+ },
2693
+ "767747424/train0": {
2694
+ "base_success": 0.25,
2695
+ "base_utility": 0.5731993420049548,
2696
+ "coverage": 0.25,
2697
+ "execute_generated": 0.25,
2698
+ "fallback_rate": 0.75,
2699
+ "hidden_chart_oracle_success": 0.25,
2700
+ "hidden_chart_oracle_utility": 0.6737305596470833,
2701
+ "outcome_ptr": 0.5,
2702
+ "predicted_margin": 0.28067674068264487,
2703
+ "proposal_oracle_success": 0.5,
2704
+ "proposal_oracle_utility": 1.0424291878007352,
2705
+ "selected_candidate_index": 7.0,
2706
+ "selected_success": 0.25,
2707
+ "selected_success_gain_over_base": 0.0,
2708
+ "selected_utility": 0.5386653137393296,
2709
+ "selected_utility_gain_over_base": -0.03453402826562524,
2710
+ "selector_regret": 0.5124941510148346,
2711
+ "success_selector_gap": 0.25,
2712
+ "success_support_gap": 0.0,
2713
+ "support_gap": 0.06562284985557199,
2714
+ "tau": 0.3473188188503786
2715
+ },
2716
+ "767747424/train1": {
2717
+ "base_success": 0.25,
2718
+ "base_utility": 0.5731993420049548,
2719
+ "coverage": 0.25,
2720
+ "execute_generated": 0.25,
2721
+ "fallback_rate": 0.75,
2722
+ "hidden_chart_oracle_success": 0.25,
2723
+ "hidden_chart_oracle_utility": 0.6737305596470833,
2724
+ "outcome_ptr": 0.5,
2725
+ "predicted_margin": 0.28220702596299274,
2726
+ "proposal_oracle_success": 0.5,
2727
+ "proposal_oracle_utility": 1.0629530912265182,
2728
+ "selected_candidate_index": 2.0,
2729
+ "selected_success": 0.25,
2730
+ "selected_success_gain_over_base": 0.0,
2731
+ "selected_utility": 0.5581304961815476,
2732
+ "selected_utility_gain_over_base": -0.015068845823407173,
2733
+ "selector_regret": 0.5076578482985497,
2734
+ "success_selector_gap": 0.25,
2735
+ "success_support_gap": 0.0,
2736
+ "support_gap": 0.045098946429789066,
2737
+ "tau": 0.3473188188503786
2738
+ },
2739
+ "767747424/train2": {
2740
+ "base_success": 0.25,
2741
+ "base_utility": 0.5731993420049548,
2742
+ "coverage": 0.25,
2743
+ "execute_generated": 0.25,
2744
+ "fallback_rate": 0.75,
2745
+ "hidden_chart_oracle_success": 0.25,
2746
+ "hidden_chart_oracle_utility": 0.6737305596470833,
2747
+ "outcome_ptr": 0.5,
2748
+ "predicted_margin": 0.28169149441007413,
2749
+ "proposal_oracle_success": 0.5,
2750
+ "proposal_oracle_utility": 1.0671670949086547,
2751
+ "selected_candidate_index": 7.75,
2752
+ "selected_success": 0.25,
2753
+ "selected_success_gain_over_base": 0.0,
2754
+ "selected_utility": 0.5799880577251315,
2755
+ "selected_utility_gain_over_base": 0.006788715720176697,
2756
+ "selector_regret": 0.48825800977647305,
2757
+ "success_selector_gap": 0.25,
2758
+ "success_support_gap": 0.0,
2759
+ "support_gap": 0.04088494274765253,
2760
+ "tau": 0.3473188188503786
2761
+ }
2762
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/metrics_by_task.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "LiftPegUpright-v1": {
3
+ "base_success": 0.2,
4
+ "base_utility": 0.9498705387115478,
5
+ "coverage": 0.0,
6
+ "execute_generated": 0.0,
7
+ "fallback_rate": 1.0,
8
+ "hidden_chart_oracle_success": 1.0,
9
+ "hidden_chart_oracle_utility": 2.0,
10
+ "outcome_ptr": 0.5333333333333333,
11
+ "predicted_margin": 0.21812074468121995,
12
+ "proposal_oracle_success": 0.6,
13
+ "proposal_oracle_utility": 1.4122399787108104,
14
+ "selected_candidate_index": 5.0,
15
+ "selected_success": 0.2,
16
+ "selected_success_gain_over_base": 0.0,
17
+ "selected_utility": 0.9498705387115478,
18
+ "selected_utility_gain_over_base": 0.0,
19
+ "selector_regret": 0.5146939317385356,
20
+ "success_selector_gap": 0.4,
21
+ "success_support_gap": 0.4,
22
+ "support_gap": 0.5877600212891897,
23
+ "tau": 0.34731881885037846
24
+ },
25
+ "PickCube-v1": {
26
+ "base_success": 0.0,
27
+ "base_utility": 0.5839591026306152,
28
+ "coverage": 0.0,
29
+ "execute_generated": 0.0,
30
+ "fallback_rate": 1.0,
31
+ "hidden_chart_oracle_success": 1.0,
32
+ "hidden_chart_oracle_utility": 2.0,
33
+ "outcome_ptr": 0.0,
34
+ "predicted_margin": 0.08871190543585684,
35
+ "proposal_oracle_success": 0.0,
36
+ "proposal_oracle_utility": 0.13425676027933756,
37
+ "selected_candidate_index": 7.666666666666667,
38
+ "selected_success": 0.0,
39
+ "selected_success_gain_over_base": 0.0,
40
+ "selected_utility": 0.5839591026306152,
41
+ "selected_utility_gain_over_base": 0.0,
42
+ "selector_regret": 0.0,
43
+ "success_selector_gap": 0.0,
44
+ "success_support_gap": 1.0,
45
+ "support_gap": 1.8657432397206624,
46
+ "tau": 0.3473188188503786
47
+ },
48
+ "PullCube-v1": {
49
+ "base_success": 0.36,
50
+ "base_utility": 0.7751704612234608,
51
+ "coverage": 0.38666666666666666,
52
+ "execute_generated": 0.38666666666666666,
53
+ "fallback_rate": 0.6133333333333333,
54
+ "hidden_chart_oracle_success": 0.52,
55
+ "hidden_chart_oracle_utility": 1.1467878860235214,
56
+ "outcome_ptr": 0.4666666666666667,
57
+ "predicted_margin": 0.33025922453409173,
58
+ "proposal_oracle_success": 0.5333333333333333,
59
+ "proposal_oracle_utility": 1.1093241602430741,
60
+ "selected_candidate_index": 3.8533333333333335,
61
+ "selected_success": 0.30666666666666664,
62
+ "selected_success_gain_over_base": -0.05333333333333334,
63
+ "selected_utility": 0.666174189314867,
64
+ "selected_utility_gain_over_base": -0.10899627190859368,
65
+ "selector_regret": 0.49621246912516653,
66
+ "success_selector_gap": 0.25333333333333335,
67
+ "success_support_gap": 0.02666666666666667,
68
+ "support_gap": 0.1075624164690574,
69
+ "tau": 0.34731881885037885
70
+ },
71
+ "PushCube-v1": {
72
+ "base_success": 0.16666666666666666,
73
+ "base_utility": 0.3924041787783305,
74
+ "coverage": 0.7222222222222222,
75
+ "execute_generated": 0.7222222222222222,
76
+ "fallback_rate": 0.2777777777777778,
77
+ "hidden_chart_oracle_success": 0.8333333333333334,
78
+ "hidden_chart_oracle_utility": 1.7015800600250561,
79
+ "outcome_ptr": 0.8333333333333334,
80
+ "predicted_margin": 0.4544794286452365,
81
+ "proposal_oracle_success": 0.8888888888888888,
82
+ "proposal_oracle_utility": 1.7857814149724112,
83
+ "selected_candidate_index": 4.555555555555555,
84
+ "selected_success": 0.2777777777777778,
85
+ "selected_success_gain_over_base": 0.1111111111111111,
86
+ "selected_utility": 0.6184621560904715,
87
+ "selected_utility_gain_over_base": 0.226057977312141,
88
+ "selector_regret": 1.1673192588819399,
89
+ "success_selector_gap": 0.6111111111111112,
90
+ "success_support_gap": 0.1111111111111111,
91
+ "support_gap": 0.21421858502758873,
92
+ "tau": 0.3473188188503784
93
+ },
94
+ "StackCube-v1": {
95
+ "base_success": 0.2727272727272727,
96
+ "base_utility": 0.7763947045261209,
97
+ "coverage": 0.2727272727272727,
98
+ "execute_generated": 0.2727272727272727,
99
+ "fallback_rate": 0.7272727272727273,
100
+ "hidden_chart_oracle_success": 1.0,
101
+ "hidden_chart_oracle_utility": 2.0,
102
+ "outcome_ptr": 0.6363636363636364,
103
+ "predicted_margin": 0.2862752255853337,
104
+ "proposal_oracle_success": 0.5151515151515151,
105
+ "proposal_oracle_utility": 1.1567951707225856,
106
+ "selected_candidate_index": 6.303030303030303,
107
+ "selected_success": 0.30303030303030304,
108
+ "selected_success_gain_over_base": 0.030303030303030304,
109
+ "selected_utility": 0.7894506924080126,
110
+ "selected_utility_gain_over_base": 0.013055987881891655,
111
+ "selector_regret": 0.37093746346054657,
112
+ "success_selector_gap": 0.21212121212121213,
113
+ "success_support_gap": 0.48484848484848486,
114
+ "support_gap": 0.8432048292774142,
115
+ "tau": 0.3473188188503786
116
+ }
117
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/report.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Nonlinear Train-Calibrated CTT Selector
2
+
3
+ Calibration rows: `432`
4
+ Fit rows: `277`
5
+ Selection rows: `155`
6
+ Eval rows: `144`
7
+ Selected model: `hgb_classifier`
8
+ Selected params: `{"l2_regularization": 0.01, "learning_rate": 0.05, "max_iter": 80, "max_leaf_nodes": 15, "random_state": 0}`
9
+ Tau: `0.347319`
10
+ Feature set: `chart_compat`
11
+ Target: `positive_margin`
12
+
13
+ The model is fit on calibration-fit rows, and model/tau are selected on held-out calibration-selection rows only. Eval outcomes are used only for reporting.
14
+
15
+ | Split | Coverage | Fallback | Base success | Selected success | Proposal oracle | OutcomePTR | Success support gap | Success selector gap |
16
+ | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
17
+ | fit | 0.4549 | 0.5451 | 0.2780 | 0.4621 | 0.5343 | 0.5343 | 0.1733 | 0.1264 |
18
+ | selection | 0.4903 | 0.5097 | 0.2774 | 0.4452 | 0.5419 | 0.5677 | 0.1806 | 0.1419 |
19
+ | eval | 0.3542 | 0.6458 | 0.2917 | 0.2847 | 0.5694 | 0.5486 | 0.2014 | 0.2986 |
20
+
21
+ This is a train-calibrated selector diagnostic over already measured candidates, not a new rollout.
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/split_hash.txt ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "calibration_target_index": {
3
+ "content_hash": "67f7b4a692f7c7e71da24378dc71e3399c5d35afc14c494869eba6087b765c42",
4
+ "retrieval_index_allowed": true,
5
+ "split": "train",
6
+ "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372"
7
+ },
8
+ "eval_target_index": {
9
+ "content_hash": "0c38d73dc9bc4ab756e12006671b5d12f1a23358b281e54b3f6a9ae3cbf6e9af",
10
+ "retrieval_index_allowed": false,
11
+ "split": "test",
12
+ "split_hash": "84151df9e88c219b76751677558cc5bbbc19082bf358cea7330158e9e403e25f"
13
+ },
14
+ "source_index": {
15
+ "content_hash": "67f7b4a692f7c7e71da24378dc71e3399c5d35afc14c494869eba6087b765c42",
16
+ "retrieval_index_allowed": true,
17
+ "split": "train",
18
+ "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372"
19
+ }
20
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/table.tex ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ % Auto-generated by scripts/eval_nonlinear_dominance_selector.py
2
+ \begin{tabular}{lrrrrrrrr}
3
+ \toprule
4
+ Rows & Coverage & Fallback & Base succ. & Selected succ. & Oracle succ. & OutcomePTR & Succ. support gap & Succ. selector gap \\
5
+ \midrule
6
+ 144 & 0.3542 & 0.6458 & 0.2917 & 0.2847 & 0.5694 & 0.5486 & 0.2014 & 0.2986 \\
7
+ \bottomrule
8
+ \end{tabular}
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_positive_envclip_k16_train_to_test/train.log ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ trained nonlinear dominance selector on calibration-fit rows only
2
+ selected_model_type=hgb_classifier
3
+ selected_model_params={"l2_regularization": 0.01, "learning_rate": 0.05, "max_iter": 80, "max_leaf_nodes": 15, "random_state": 0}
4
+ tau=0.347319
5
+ calibration_input=runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/command.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ python scripts/eval_nonlinear_dominance_selector.py --calibration-input runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json --calibration-target-index data/cil_charts_rgb_refs/train/index.json --eval-input runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json --eval-target-index data/cil_charts_rgb_refs/test/index.json --source-index data/cil_charts_rgb_refs/train/index.json --checkpoint-template runs/ctt_residual_base_context_obs_seed{seed}/model.pt --out-dir runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test --feature-set chart_compat --selector-chart-feature-mode base_context_obs --target success_weighted_margin --k 16 --bootstrap-samples 1000
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/config.yaml ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ bootstrap_samples: 1000
2
+ calibration_input: runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json
3
+ calibration_target_index: data/cil_charts_rgb_refs/train/index.json
4
+ checkpoint_template: runs/ctt_residual_base_context_obs_seed{seed}/model.pt
5
+ eval_input: runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json
6
+ eval_target_index: data/cil_charts_rgb_refs/test/index.json
7
+ feature_set: chart_compat
8
+ k: 16
9
+ model_types: hgb_classifier,hgb_regressor,rf_regressor
10
+ out_dir: runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test
11
+ seed: 0
12
+ selection_frac: 0.35
13
+ selector_chart_feature_mode: base_context_obs
14
+ source_index: data/cil_charts_rgb_refs/train/index.json
15
+ target: success_weighted_margin
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/data_hash.txt ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "calibration_input": "d79e275b896eb98caa128470d4614b76960cab12407faa157c583224be7abe80",
3
+ "calibration_target_index": "67973c50072aa87a3b97533b6292b08a8b4d88a95571dc22e91b6e8fd19e0d85",
4
+ "eval_input": "541d5a928da5f7ee04ca3dea570e04208830dffde431edfe5b5e38b692634045",
5
+ "eval_target_index": "dee1b5225485df6624747dd4c5074bcf514685b54017d05e698d2051638fbf36",
6
+ "source_index": "67973c50072aa87a3b97533b6292b08a8b4d88a95571dc22e91b6e8fd19e0d85"
7
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/eval.log ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ selected model/tau on held-out calibration-selection rows; evaluated on held-out measured rollout rows
2
+ eval_input=runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json
3
+ num_eval_rows=144
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/git_hash.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ 14e81292e2004925f81e0bb4810f6f144bfd87b7
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/metrics.json ADDED
The diff for this file is too large to render. See raw diff
 
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/metrics_by_seed.json ADDED
@@ -0,0 +1,2762 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "1073161778/train0": {
3
+ "base_success": 1.0,
4
+ "base_utility": 2.0,
5
+ "coverage": 0.0,
6
+ "execute_generated": 0.0,
7
+ "fallback_rate": 1.0,
8
+ "hidden_chart_oracle_success": 1.0,
9
+ "hidden_chart_oracle_utility": 2.0,
10
+ "outcome_ptr": 0.0,
11
+ "predicted_margin": 0.28355827608242967,
12
+ "proposal_oracle_success": 1.0,
13
+ "proposal_oracle_utility": 2.0,
14
+ "selected_candidate_index": 6.0,
15
+ "selected_success": 1.0,
16
+ "selected_success_gain_over_base": 0.0,
17
+ "selected_utility": 2.0,
18
+ "selected_utility_gain_over_base": 0.0,
19
+ "selector_regret": 0.0,
20
+ "success_selector_gap": 0.0,
21
+ "success_support_gap": 0.0,
22
+ "support_gap": 0.0,
23
+ "tau": 0.3473188188503786
24
+ },
25
+ "1073161778/train1": {
26
+ "base_success": 1.0,
27
+ "base_utility": 2.0,
28
+ "coverage": 0.0,
29
+ "execute_generated": 0.0,
30
+ "fallback_rate": 1.0,
31
+ "hidden_chart_oracle_success": 1.0,
32
+ "hidden_chart_oracle_utility": 2.0,
33
+ "outcome_ptr": 0.0,
34
+ "predicted_margin": 0.3337636013451598,
35
+ "proposal_oracle_success": 1.0,
36
+ "proposal_oracle_utility": 2.0,
37
+ "selected_candidate_index": 7.0,
38
+ "selected_success": 1.0,
39
+ "selected_success_gain_over_base": 0.0,
40
+ "selected_utility": 2.0,
41
+ "selected_utility_gain_over_base": 0.0,
42
+ "selector_regret": 0.0,
43
+ "success_selector_gap": 0.0,
44
+ "success_support_gap": 0.0,
45
+ "support_gap": 0.0,
46
+ "tau": 0.3473188188503786
47
+ },
48
+ "1073161778/train2": {
49
+ "base_success": 1.0,
50
+ "base_utility": 2.0,
51
+ "coverage": 0.0,
52
+ "execute_generated": 0.0,
53
+ "fallback_rate": 1.0,
54
+ "hidden_chart_oracle_success": 1.0,
55
+ "hidden_chart_oracle_utility": 2.0,
56
+ "outcome_ptr": 0.0,
57
+ "predicted_margin": 0.29522744379276233,
58
+ "proposal_oracle_success": 1.0,
59
+ "proposal_oracle_utility": 2.0,
60
+ "selected_candidate_index": 3.0,
61
+ "selected_success": 1.0,
62
+ "selected_success_gain_over_base": 0.0,
63
+ "selected_utility": 2.0,
64
+ "selected_utility_gain_over_base": 0.0,
65
+ "selector_regret": 0.0,
66
+ "success_selector_gap": 0.0,
67
+ "success_support_gap": 0.0,
68
+ "support_gap": 0.0,
69
+ "tau": 0.3473188188503786
70
+ },
71
+ "1113371261/train0": {
72
+ "base_success": 0.0,
73
+ "base_utility": 0.15842342376708984,
74
+ "coverage": 0.0,
75
+ "execute_generated": 0.0,
76
+ "fallback_rate": 1.0,
77
+ "hidden_chart_oracle_success": 1.0,
78
+ "hidden_chart_oracle_utility": 2.0,
79
+ "outcome_ptr": 0.0,
80
+ "predicted_margin": 0.32662525863909614,
81
+ "proposal_oracle_success": 0.0,
82
+ "proposal_oracle_utility": 0.10274861752986908,
83
+ "selected_candidate_index": 0.0,
84
+ "selected_success": 0.0,
85
+ "selected_success_gain_over_base": 0.0,
86
+ "selected_utility": 0.15842342376708984,
87
+ "selected_utility_gain_over_base": 0.0,
88
+ "selector_regret": 0.0,
89
+ "success_selector_gap": 0.0,
90
+ "success_support_gap": 1.0,
91
+ "support_gap": 1.897251382470131,
92
+ "tau": 0.3473188188503786
93
+ },
94
+ "1113371261/train1": {
95
+ "base_success": 0.0,
96
+ "base_utility": 0.15842342376708984,
97
+ "coverage": 0.0,
98
+ "execute_generated": 0.0,
99
+ "fallback_rate": 1.0,
100
+ "hidden_chart_oracle_success": 1.0,
101
+ "hidden_chart_oracle_utility": 2.0,
102
+ "outcome_ptr": 1.0,
103
+ "predicted_margin": 0.27481853311866083,
104
+ "proposal_oracle_success": 0.0,
105
+ "proposal_oracle_utility": 0.18093222379684448,
106
+ "selected_candidate_index": 4.0,
107
+ "selected_success": 0.0,
108
+ "selected_success_gain_over_base": 0.0,
109
+ "selected_utility": 0.15842342376708984,
110
+ "selected_utility_gain_over_base": 0.0,
111
+ "selector_regret": 0.02250880002975464,
112
+ "success_selector_gap": 0.0,
113
+ "success_support_gap": 1.0,
114
+ "support_gap": 1.8190677762031555,
115
+ "tau": 0.3473188188503786
116
+ },
117
+ "1113371261/train2": {
118
+ "base_success": 0.0,
119
+ "base_utility": 0.15842342376708984,
120
+ "coverage": 0.0,
121
+ "execute_generated": 0.0,
122
+ "fallback_rate": 1.0,
123
+ "hidden_chart_oracle_success": 1.0,
124
+ "hidden_chart_oracle_utility": 2.0,
125
+ "outcome_ptr": 1.0,
126
+ "predicted_margin": 0.3131801253335776,
127
+ "proposal_oracle_success": 0.0,
128
+ "proposal_oracle_utility": 0.19527167081832886,
129
+ "selected_candidate_index": 2.0,
130
+ "selected_success": 0.0,
131
+ "selected_success_gain_over_base": 0.0,
132
+ "selected_utility": 0.15842342376708984,
133
+ "selected_utility_gain_over_base": 0.0,
134
+ "selector_regret": 0.036848247051239014,
135
+ "success_selector_gap": 0.0,
136
+ "success_support_gap": 1.0,
137
+ "support_gap": 1.8047283291816711,
138
+ "tau": 0.3473188188503786
139
+ },
140
+ "1151410346/train0": {
141
+ "base_success": 0.0,
142
+ "base_utility": 0.2260012924671173,
143
+ "coverage": 0.0,
144
+ "execute_generated": 0.0,
145
+ "fallback_rate": 1.0,
146
+ "hidden_chart_oracle_success": 1.0,
147
+ "hidden_chart_oracle_utility": 2.0,
148
+ "outcome_ptr": 1.0,
149
+ "predicted_margin": 0.20532758727946646,
150
+ "proposal_oracle_success": 0.0,
151
+ "proposal_oracle_utility": 0.8663521409034729,
152
+ "selected_candidate_index": 7.0,
153
+ "selected_success": 0.0,
154
+ "selected_success_gain_over_base": 0.0,
155
+ "selected_utility": 0.2260012924671173,
156
+ "selected_utility_gain_over_base": 0.0,
157
+ "selector_regret": 0.6403508484363556,
158
+ "success_selector_gap": 0.0,
159
+ "success_support_gap": 1.0,
160
+ "support_gap": 1.133647859096527,
161
+ "tau": 0.3473188188503786
162
+ },
163
+ "1151410346/train1": {
164
+ "base_success": 0.0,
165
+ "base_utility": 0.2260012924671173,
166
+ "coverage": 0.0,
167
+ "execute_generated": 0.0,
168
+ "fallback_rate": 1.0,
169
+ "hidden_chart_oracle_success": 1.0,
170
+ "hidden_chart_oracle_utility": 2.0,
171
+ "outcome_ptr": 1.0,
172
+ "predicted_margin": 0.21665335322366186,
173
+ "proposal_oracle_success": 1.0,
174
+ "proposal_oracle_utility": 2.0,
175
+ "selected_candidate_index": 9.0,
176
+ "selected_success": 0.0,
177
+ "selected_success_gain_over_base": 0.0,
178
+ "selected_utility": 0.2260012924671173,
179
+ "selected_utility_gain_over_base": 0.0,
180
+ "selector_regret": 1.7739987075328827,
181
+ "success_selector_gap": 1.0,
182
+ "success_support_gap": 0.0,
183
+ "support_gap": 0.0,
184
+ "tau": 0.3473188188503786
185
+ },
186
+ "1151410346/train2": {
187
+ "base_success": 0.0,
188
+ "base_utility": 0.2260012924671173,
189
+ "coverage": 0.0,
190
+ "execute_generated": 0.0,
191
+ "fallback_rate": 1.0,
192
+ "hidden_chart_oracle_success": 1.0,
193
+ "hidden_chart_oracle_utility": 2.0,
194
+ "outcome_ptr": 1.0,
195
+ "predicted_margin": 0.22061568738801626,
196
+ "proposal_oracle_success": 0.0,
197
+ "proposal_oracle_utility": 0.5522047281265259,
198
+ "selected_candidate_index": 10.0,
199
+ "selected_success": 0.0,
200
+ "selected_success_gain_over_base": 0.0,
201
+ "selected_utility": 0.2260012924671173,
202
+ "selected_utility_gain_over_base": 0.0,
203
+ "selector_regret": 0.32620343565940857,
204
+ "success_selector_gap": 0.0,
205
+ "success_support_gap": 1.0,
206
+ "support_gap": 1.4477952718734741,
207
+ "tau": 0.3473188188503786
208
+ },
209
+ "1261930456/train0": {
210
+ "base_success": 1.0,
211
+ "base_utility": 2.0,
212
+ "coverage": 1.0,
213
+ "execute_generated": 1.0,
214
+ "fallback_rate": 0.0,
215
+ "hidden_chart_oracle_success": 1.0,
216
+ "hidden_chart_oracle_utility": 2.0,
217
+ "outcome_ptr": 0.0,
218
+ "predicted_margin": 0.5566373654374286,
219
+ "proposal_oracle_success": 1.0,
220
+ "proposal_oracle_utility": 2.0,
221
+ "selected_candidate_index": 5.0,
222
+ "selected_success": 1.0,
223
+ "selected_success_gain_over_base": 0.0,
224
+ "selected_utility": 2.0,
225
+ "selected_utility_gain_over_base": 0.0,
226
+ "selector_regret": 0.0,
227
+ "success_selector_gap": 0.0,
228
+ "success_support_gap": 0.0,
229
+ "support_gap": 0.0,
230
+ "tau": 0.3473188188503786
231
+ },
232
+ "1261930456/train1": {
233
+ "base_success": 1.0,
234
+ "base_utility": 2.0,
235
+ "coverage": 1.0,
236
+ "execute_generated": 1.0,
237
+ "fallback_rate": 0.0,
238
+ "hidden_chart_oracle_success": 1.0,
239
+ "hidden_chart_oracle_utility": 2.0,
240
+ "outcome_ptr": 0.0,
241
+ "predicted_margin": 0.5940825716503084,
242
+ "proposal_oracle_success": 1.0,
243
+ "proposal_oracle_utility": 2.0,
244
+ "selected_candidate_index": 0.0,
245
+ "selected_success": 1.0,
246
+ "selected_success_gain_over_base": 0.0,
247
+ "selected_utility": 2.0,
248
+ "selected_utility_gain_over_base": 0.0,
249
+ "selector_regret": 0.0,
250
+ "success_selector_gap": 0.0,
251
+ "success_support_gap": 0.0,
252
+ "support_gap": 0.0,
253
+ "tau": 0.3473188188503786
254
+ },
255
+ "1261930456/train2": {
256
+ "base_success": 1.0,
257
+ "base_utility": 2.0,
258
+ "coverage": 1.0,
259
+ "execute_generated": 1.0,
260
+ "fallback_rate": 0.0,
261
+ "hidden_chart_oracle_success": 1.0,
262
+ "hidden_chart_oracle_utility": 2.0,
263
+ "outcome_ptr": 0.0,
264
+ "predicted_margin": 0.5877928871421093,
265
+ "proposal_oracle_success": 1.0,
266
+ "proposal_oracle_utility": 2.0,
267
+ "selected_candidate_index": 1.0,
268
+ "selected_success": 0.0,
269
+ "selected_success_gain_over_base": -1.0,
270
+ "selected_utility": 0.04538704454898834,
271
+ "selected_utility_gain_over_base": -1.9546129554510117,
272
+ "selector_regret": 1.9546129554510117,
273
+ "success_selector_gap": 1.0,
274
+ "success_support_gap": 0.0,
275
+ "support_gap": 0.0,
276
+ "tau": 0.3473188188503786
277
+ },
278
+ "1324133122/train0": {
279
+ "base_success": 0.0,
280
+ "base_utility": 0.09220491349697113,
281
+ "coverage": 0.0,
282
+ "execute_generated": 0.0,
283
+ "fallback_rate": 1.0,
284
+ "hidden_chart_oracle_success": 1.0,
285
+ "hidden_chart_oracle_utility": 2.0,
286
+ "outcome_ptr": 0.0,
287
+ "predicted_margin": 0.1953526650841092,
288
+ "proposal_oracle_success": 0.0,
289
+ "proposal_oracle_utility": 0.054989203810691833,
290
+ "selected_candidate_index": 0.0,
291
+ "selected_success": 0.0,
292
+ "selected_success_gain_over_base": 0.0,
293
+ "selected_utility": 0.09220491349697113,
294
+ "selected_utility_gain_over_base": 0.0,
295
+ "selector_regret": 0.0,
296
+ "success_selector_gap": 0.0,
297
+ "success_support_gap": 1.0,
298
+ "support_gap": 1.9450107961893082,
299
+ "tau": 0.3473188188503786
300
+ },
301
+ "1324133122/train1": {
302
+ "base_success": 0.0,
303
+ "base_utility": 0.09220491349697113,
304
+ "coverage": 0.0,
305
+ "execute_generated": 0.0,
306
+ "fallback_rate": 1.0,
307
+ "hidden_chart_oracle_success": 1.0,
308
+ "hidden_chart_oracle_utility": 2.0,
309
+ "outcome_ptr": 1.0,
310
+ "predicted_margin": 0.10311308876908776,
311
+ "proposal_oracle_success": 0.0,
312
+ "proposal_oracle_utility": 0.1324586570262909,
313
+ "selected_candidate_index": 5.0,
314
+ "selected_success": 0.0,
315
+ "selected_success_gain_over_base": 0.0,
316
+ "selected_utility": 0.09220491349697113,
317
+ "selected_utility_gain_over_base": 0.0,
318
+ "selector_regret": 0.04025374352931976,
319
+ "success_selector_gap": 0.0,
320
+ "success_support_gap": 1.0,
321
+ "support_gap": 1.867541342973709,
322
+ "tau": 0.3473188188503786
323
+ },
324
+ "1324133122/train2": {
325
+ "base_success": 0.0,
326
+ "base_utility": 0.09220491349697113,
327
+ "coverage": 0.0,
328
+ "execute_generated": 0.0,
329
+ "fallback_rate": 1.0,
330
+ "hidden_chart_oracle_success": 1.0,
331
+ "hidden_chart_oracle_utility": 2.0,
332
+ "outcome_ptr": 1.0,
333
+ "predicted_margin": 0.17894998490871183,
334
+ "proposal_oracle_success": 0.0,
335
+ "proposal_oracle_utility": 0.19291165471076965,
336
+ "selected_candidate_index": 7.0,
337
+ "selected_success": 0.0,
338
+ "selected_success_gain_over_base": 0.0,
339
+ "selected_utility": 0.09220491349697113,
340
+ "selected_utility_gain_over_base": 0.0,
341
+ "selector_regret": 0.10070674121379852,
342
+ "success_selector_gap": 0.0,
343
+ "success_support_gap": 1.0,
344
+ "support_gap": 1.8070883452892303,
345
+ "tau": 0.3473188188503786
346
+ },
347
+ "1324642411/train0": {
348
+ "base_success": 0.0,
349
+ "base_utility": 0.10647052526473999,
350
+ "coverage": 1.0,
351
+ "execute_generated": 1.0,
352
+ "fallback_rate": 0.0,
353
+ "hidden_chart_oracle_success": 1.0,
354
+ "hidden_chart_oracle_utility": 2.0,
355
+ "outcome_ptr": 1.0,
356
+ "predicted_margin": 0.5244761772076728,
357
+ "proposal_oracle_success": 0.0,
358
+ "proposal_oracle_utility": 0.14406004548072815,
359
+ "selected_candidate_index": 3.0,
360
+ "selected_success": 0.0,
361
+ "selected_success_gain_over_base": 0.0,
362
+ "selected_utility": 0.061532557010650635,
363
+ "selected_utility_gain_over_base": -0.044937968254089355,
364
+ "selector_regret": 0.08252748847007751,
365
+ "success_selector_gap": 0.0,
366
+ "success_support_gap": 1.0,
367
+ "support_gap": 1.8559399545192719,
368
+ "tau": 0.3473188188503786
369
+ },
370
+ "1324642411/train1": {
371
+ "base_success": 0.0,
372
+ "base_utility": 0.10647052526473999,
373
+ "coverage": 1.0,
374
+ "execute_generated": 1.0,
375
+ "fallback_rate": 0.0,
376
+ "hidden_chart_oracle_success": 1.0,
377
+ "hidden_chart_oracle_utility": 2.0,
378
+ "outcome_ptr": 1.0,
379
+ "predicted_margin": 0.48070996484280953,
380
+ "proposal_oracle_success": 1.0,
381
+ "proposal_oracle_utility": 2.0,
382
+ "selected_candidate_index": 9.0,
383
+ "selected_success": 0.0,
384
+ "selected_success_gain_over_base": 0.0,
385
+ "selected_utility": 0.1280023753643036,
386
+ "selected_utility_gain_over_base": 0.0215318500995636,
387
+ "selector_regret": 1.8719976246356964,
388
+ "success_selector_gap": 1.0,
389
+ "success_support_gap": 0.0,
390
+ "support_gap": 0.0,
391
+ "tau": 0.3473188188503786
392
+ },
393
+ "1324642411/train2": {
394
+ "base_success": 0.0,
395
+ "base_utility": 0.10647052526473999,
396
+ "coverage": 1.0,
397
+ "execute_generated": 1.0,
398
+ "fallback_rate": 0.0,
399
+ "hidden_chart_oracle_success": 1.0,
400
+ "hidden_chart_oracle_utility": 2.0,
401
+ "outcome_ptr": 1.0,
402
+ "predicted_margin": 0.4884411626074505,
403
+ "proposal_oracle_success": 1.0,
404
+ "proposal_oracle_utility": 2.0,
405
+ "selected_candidate_index": 3.0,
406
+ "selected_success": 1.0,
407
+ "selected_success_gain_over_base": 1.0,
408
+ "selected_utility": 2.0,
409
+ "selected_utility_gain_over_base": 1.89352947473526,
410
+ "selector_regret": 0.0,
411
+ "success_selector_gap": 0.0,
412
+ "success_support_gap": 0.0,
413
+ "support_gap": 0.0,
414
+ "tau": 0.3473188188503786
415
+ },
416
+ "1375492786/train0": {
417
+ "base_success": 1.0,
418
+ "base_utility": 2.0,
419
+ "coverage": 0.0,
420
+ "execute_generated": 0.0,
421
+ "fallback_rate": 1.0,
422
+ "hidden_chart_oracle_success": 1.0,
423
+ "hidden_chart_oracle_utility": 2.0,
424
+ "outcome_ptr": 0.0,
425
+ "predicted_margin": 0.16043123876023302,
426
+ "proposal_oracle_success": 1.0,
427
+ "proposal_oracle_utility": 2.0,
428
+ "selected_candidate_index": 15.0,
429
+ "selected_success": 1.0,
430
+ "selected_success_gain_over_base": 0.0,
431
+ "selected_utility": 2.0,
432
+ "selected_utility_gain_over_base": 0.0,
433
+ "selector_regret": 0.0,
434
+ "success_selector_gap": 0.0,
435
+ "success_support_gap": 0.0,
436
+ "support_gap": 0.0,
437
+ "tau": 0.3473188188503786
438
+ },
439
+ "1375492786/train1": {
440
+ "base_success": 1.0,
441
+ "base_utility": 2.0,
442
+ "coverage": 0.0,
443
+ "execute_generated": 0.0,
444
+ "fallback_rate": 1.0,
445
+ "hidden_chart_oracle_success": 1.0,
446
+ "hidden_chart_oracle_utility": 2.0,
447
+ "outcome_ptr": 0.0,
448
+ "predicted_margin": 0.17011388186485987,
449
+ "proposal_oracle_success": 1.0,
450
+ "proposal_oracle_utility": 2.0,
451
+ "selected_candidate_index": 10.0,
452
+ "selected_success": 1.0,
453
+ "selected_success_gain_over_base": 0.0,
454
+ "selected_utility": 2.0,
455
+ "selected_utility_gain_over_base": 0.0,
456
+ "selector_regret": 0.0,
457
+ "success_selector_gap": 0.0,
458
+ "success_support_gap": 0.0,
459
+ "support_gap": 0.0,
460
+ "tau": 0.3473188188503786
461
+ },
462
+ "1375492786/train2": {
463
+ "base_success": 1.0,
464
+ "base_utility": 2.0,
465
+ "coverage": 0.0,
466
+ "execute_generated": 0.0,
467
+ "fallback_rate": 1.0,
468
+ "hidden_chart_oracle_success": 1.0,
469
+ "hidden_chart_oracle_utility": 2.0,
470
+ "outcome_ptr": 0.0,
471
+ "predicted_margin": 0.15542103752775793,
472
+ "proposal_oracle_success": 1.0,
473
+ "proposal_oracle_utility": 2.0,
474
+ "selected_candidate_index": 15.0,
475
+ "selected_success": 1.0,
476
+ "selected_success_gain_over_base": 0.0,
477
+ "selected_utility": 2.0,
478
+ "selected_utility_gain_over_base": 0.0,
479
+ "selector_regret": 0.0,
480
+ "success_selector_gap": 0.0,
481
+ "success_support_gap": 0.0,
482
+ "support_gap": 0.0,
483
+ "tau": 0.3473188188503786
484
+ },
485
+ "141984615/train0": {
486
+ "base_success": 0.0,
487
+ "base_utility": 0.10283424705266953,
488
+ "coverage": 1.0,
489
+ "execute_generated": 1.0,
490
+ "fallback_rate": 0.0,
491
+ "hidden_chart_oracle_success": 0.0,
492
+ "hidden_chart_oracle_utility": 0.2857772707939148,
493
+ "outcome_ptr": 0.0,
494
+ "predicted_margin": 0.3988789611600871,
495
+ "proposal_oracle_success": 0.0,
496
+ "proposal_oracle_utility": 0.015177905559539795,
497
+ "selected_candidate_index": 1.0,
498
+ "selected_success": 0.0,
499
+ "selected_success_gain_over_base": 0.0,
500
+ "selected_utility": 0.0042244791984558105,
501
+ "selected_utility_gain_over_base": -0.09860976785421371,
502
+ "selector_regret": 0.010953426361083984,
503
+ "success_selector_gap": 0.0,
504
+ "success_support_gap": 0.0,
505
+ "support_gap": 0.270599365234375,
506
+ "tau": 0.3473188188503786
507
+ },
508
+ "141984615/train1": {
509
+ "base_success": 0.0,
510
+ "base_utility": 0.10283424705266953,
511
+ "coverage": 1.0,
512
+ "execute_generated": 1.0,
513
+ "fallback_rate": 0.0,
514
+ "hidden_chart_oracle_success": 0.0,
515
+ "hidden_chart_oracle_utility": 0.2857772707939148,
516
+ "outcome_ptr": 0.0,
517
+ "predicted_margin": 0.37538333059509166,
518
+ "proposal_oracle_success": 0.0,
519
+ "proposal_oracle_utility": 0.037011127918958664,
520
+ "selected_candidate_index": 6.0,
521
+ "selected_success": 0.0,
522
+ "selected_success_gain_over_base": 0.0,
523
+ "selected_utility": 0.035802166908979416,
524
+ "selected_utility_gain_over_base": -0.06703208014369011,
525
+ "selector_regret": 0.001208961009979248,
526
+ "success_selector_gap": 0.0,
527
+ "success_support_gap": 0.0,
528
+ "support_gap": 0.24876614287495613,
529
+ "tau": 0.3473188188503786
530
+ },
531
+ "141984615/train2": {
532
+ "base_success": 0.0,
533
+ "base_utility": 0.10283424705266953,
534
+ "coverage": 0.0,
535
+ "execute_generated": 0.0,
536
+ "fallback_rate": 1.0,
537
+ "hidden_chart_oracle_success": 0.0,
538
+ "hidden_chart_oracle_utility": 0.2857772707939148,
539
+ "outcome_ptr": 0.0,
540
+ "predicted_margin": 0.30448894670641713,
541
+ "proposal_oracle_success": 0.0,
542
+ "proposal_oracle_utility": 0.058951716870069504,
543
+ "selected_candidate_index": 4.0,
544
+ "selected_success": 0.0,
545
+ "selected_success_gain_over_base": 0.0,
546
+ "selected_utility": 0.10283424705266953,
547
+ "selected_utility_gain_over_base": 0.0,
548
+ "selector_regret": 0.0,
549
+ "success_selector_gap": 0.0,
550
+ "success_support_gap": 0.0,
551
+ "support_gap": 0.2268255539238453,
552
+ "tau": 0.3473188188503786
553
+ },
554
+ "1461069713/train0": {
555
+ "base_success": 0.0,
556
+ "base_utility": 0.0939030647277832,
557
+ "coverage": 0.0,
558
+ "execute_generated": 0.0,
559
+ "fallback_rate": 1.0,
560
+ "hidden_chart_oracle_success": 0.0,
561
+ "hidden_chart_oracle_utility": 0.1593424528837204,
562
+ "outcome_ptr": 0.0,
563
+ "predicted_margin": 0.3050981300072402,
564
+ "proposal_oracle_success": 0.0,
565
+ "proposal_oracle_utility": 0.06019836664199829,
566
+ "selected_candidate_index": 2.0,
567
+ "selected_success": 0.0,
568
+ "selected_success_gain_over_base": 0.0,
569
+ "selected_utility": 0.0939030647277832,
570
+ "selected_utility_gain_over_base": 0.0,
571
+ "selector_regret": 0.0,
572
+ "success_selector_gap": 0.0,
573
+ "success_support_gap": 0.0,
574
+ "support_gap": 0.0991440862417221,
575
+ "tau": 0.3473188188503786
576
+ },
577
+ "1461069713/train1": {
578
+ "base_success": 0.0,
579
+ "base_utility": 0.0939030647277832,
580
+ "coverage": 1.0,
581
+ "execute_generated": 1.0,
582
+ "fallback_rate": 0.0,
583
+ "hidden_chart_oracle_success": 0.0,
584
+ "hidden_chart_oracle_utility": 0.1593424528837204,
585
+ "outcome_ptr": 0.0,
586
+ "predicted_margin": 0.7484074094930177,
587
+ "proposal_oracle_success": 0.0,
588
+ "proposal_oracle_utility": 0.016585012897849083,
589
+ "selected_candidate_index": 0.0,
590
+ "selected_success": 0.0,
591
+ "selected_success_gain_over_base": 0.0,
592
+ "selected_utility": 0.009687979705631733,
593
+ "selected_utility_gain_over_base": -0.08421508502215147,
594
+ "selector_regret": 0.00689703319221735,
595
+ "success_selector_gap": 0.0,
596
+ "success_support_gap": 0.0,
597
+ "support_gap": 0.14275743998587132,
598
+ "tau": 0.3473188188503786
599
+ },
600
+ "1461069713/train2": {
601
+ "base_success": 0.0,
602
+ "base_utility": 0.0939030647277832,
603
+ "coverage": 0.0,
604
+ "execute_generated": 0.0,
605
+ "fallback_rate": 1.0,
606
+ "hidden_chart_oracle_success": 0.0,
607
+ "hidden_chart_oracle_utility": 0.1593424528837204,
608
+ "outcome_ptr": 1.0,
609
+ "predicted_margin": 0.2910706981901193,
610
+ "proposal_oracle_success": 0.0,
611
+ "proposal_oracle_utility": 0.09997648000717163,
612
+ "selected_candidate_index": 5.0,
613
+ "selected_success": 0.0,
614
+ "selected_success_gain_over_base": 0.0,
615
+ "selected_utility": 0.0939030647277832,
616
+ "selected_utility_gain_over_base": 0.0,
617
+ "selector_regret": 0.006073415279388428,
618
+ "success_selector_gap": 0.0,
619
+ "success_support_gap": 0.0,
620
+ "support_gap": 0.05936597287654877,
621
+ "tau": 0.3473188188503786
622
+ },
623
+ "1480856810/train0": {
624
+ "base_success": 0.0,
625
+ "base_utility": 0.08703407645225525,
626
+ "coverage": 0.0,
627
+ "execute_generated": 0.0,
628
+ "fallback_rate": 1.0,
629
+ "hidden_chart_oracle_success": 0.0,
630
+ "hidden_chart_oracle_utility": 0.20948036015033722,
631
+ "outcome_ptr": 1.0,
632
+ "predicted_margin": 0.29221027147425893,
633
+ "proposal_oracle_success": 1.0,
634
+ "proposal_oracle_utility": 2.0,
635
+ "selected_candidate_index": 2.0,
636
+ "selected_success": 0.0,
637
+ "selected_success_gain_over_base": 0.0,
638
+ "selected_utility": 0.08703407645225525,
639
+ "selected_utility_gain_over_base": 0.0,
640
+ "selector_regret": 1.9129659235477448,
641
+ "success_selector_gap": 1.0,
642
+ "success_support_gap": 0.0,
643
+ "support_gap": 0.0,
644
+ "tau": 0.3473188188503786
645
+ },
646
+ "1480856810/train1": {
647
+ "base_success": 0.0,
648
+ "base_utility": 0.08703407645225525,
649
+ "coverage": 1.0,
650
+ "execute_generated": 1.0,
651
+ "fallback_rate": 0.0,
652
+ "hidden_chart_oracle_success": 0.0,
653
+ "hidden_chart_oracle_utility": 0.20948036015033722,
654
+ "outcome_ptr": 1.0,
655
+ "predicted_margin": 0.43921922377953376,
656
+ "proposal_oracle_success": 1.0,
657
+ "proposal_oracle_utility": 2.0,
658
+ "selected_candidate_index": 13.0,
659
+ "selected_success": 0.0,
660
+ "selected_success_gain_over_base": 0.0,
661
+ "selected_utility": 0.20681259036064148,
662
+ "selected_utility_gain_over_base": 0.11977851390838623,
663
+ "selector_regret": 1.7931874096393585,
664
+ "success_selector_gap": 1.0,
665
+ "success_support_gap": 0.0,
666
+ "support_gap": 0.0,
667
+ "tau": 0.3473188188503786
668
+ },
669
+ "1480856810/train2": {
670
+ "base_success": 0.0,
671
+ "base_utility": 0.08703407645225525,
672
+ "coverage": 0.0,
673
+ "execute_generated": 0.0,
674
+ "fallback_rate": 1.0,
675
+ "hidden_chart_oracle_success": 0.0,
676
+ "hidden_chart_oracle_utility": 0.20948036015033722,
677
+ "outcome_ptr": 1.0,
678
+ "predicted_margin": 0.29221027147425893,
679
+ "proposal_oracle_success": 1.0,
680
+ "proposal_oracle_utility": 2.0,
681
+ "selected_candidate_index": 1.0,
682
+ "selected_success": 0.0,
683
+ "selected_success_gain_over_base": 0.0,
684
+ "selected_utility": 0.08703407645225525,
685
+ "selected_utility_gain_over_base": 0.0,
686
+ "selector_regret": 1.9129659235477448,
687
+ "success_selector_gap": 1.0,
688
+ "success_support_gap": 0.0,
689
+ "support_gap": 0.0,
690
+ "tau": 0.3473188188503786
691
+ },
692
+ "1509980664/train0": {
693
+ "base_success": 0.0,
694
+ "base_utility": 0.8334100246429443,
695
+ "coverage": 1.0,
696
+ "execute_generated": 1.0,
697
+ "fallback_rate": 0.0,
698
+ "hidden_chart_oracle_success": 1.0,
699
+ "hidden_chart_oracle_utility": 2.0,
700
+ "outcome_ptr": 1.0,
701
+ "predicted_margin": 0.5870819789979614,
702
+ "proposal_oracle_success": 1.0,
703
+ "proposal_oracle_utility": 2.0,
704
+ "selected_candidate_index": 10.0,
705
+ "selected_success": 1.0,
706
+ "selected_success_gain_over_base": 1.0,
707
+ "selected_utility": 2.0,
708
+ "selected_utility_gain_over_base": 1.1665899753570557,
709
+ "selector_regret": 0.0,
710
+ "success_selector_gap": 0.0,
711
+ "success_support_gap": 0.0,
712
+ "support_gap": 0.0,
713
+ "tau": 0.3473188188503786
714
+ },
715
+ "1509980664/train1": {
716
+ "base_success": 0.0,
717
+ "base_utility": 0.8334100246429443,
718
+ "coverage": 1.0,
719
+ "execute_generated": 1.0,
720
+ "fallback_rate": 0.0,
721
+ "hidden_chart_oracle_success": 1.0,
722
+ "hidden_chart_oracle_utility": 2.0,
723
+ "outcome_ptr": 1.0,
724
+ "predicted_margin": 0.6097239173416052,
725
+ "proposal_oracle_success": 1.0,
726
+ "proposal_oracle_utility": 2.0,
727
+ "selected_candidate_index": 4.0,
728
+ "selected_success": 0.0,
729
+ "selected_success_gain_over_base": 0.0,
730
+ "selected_utility": 0.21804393827915192,
731
+ "selected_utility_gain_over_base": -0.6153660863637924,
732
+ "selector_regret": 1.781956061720848,
733
+ "success_selector_gap": 1.0,
734
+ "success_support_gap": 0.0,
735
+ "support_gap": 0.0,
736
+ "tau": 0.3473188188503786
737
+ },
738
+ "1509980664/train2": {
739
+ "base_success": 0.0,
740
+ "base_utility": 0.8334100246429443,
741
+ "coverage": 1.0,
742
+ "execute_generated": 1.0,
743
+ "fallback_rate": 0.0,
744
+ "hidden_chart_oracle_success": 1.0,
745
+ "hidden_chart_oracle_utility": 2.0,
746
+ "outcome_ptr": 1.0,
747
+ "predicted_margin": 0.5974323487206175,
748
+ "proposal_oracle_success": 1.0,
749
+ "proposal_oracle_utility": 2.0,
750
+ "selected_candidate_index": 12.0,
751
+ "selected_success": 0.0,
752
+ "selected_success_gain_over_base": 0.0,
753
+ "selected_utility": 0.8189421892166138,
754
+ "selected_utility_gain_over_base": -0.014467835426330566,
755
+ "selector_regret": 1.1810578107833862,
756
+ "success_selector_gap": 1.0,
757
+ "success_support_gap": 0.0,
758
+ "support_gap": 0.0,
759
+ "tau": 0.3473188188503786
760
+ },
761
+ "1522272686/train0": {
762
+ "base_success": 0.0,
763
+ "base_utility": 0.5839591026306152,
764
+ "coverage": 0.0,
765
+ "execute_generated": 0.0,
766
+ "fallback_rate": 1.0,
767
+ "hidden_chart_oracle_success": 1.0,
768
+ "hidden_chart_oracle_utility": 2.0,
769
+ "outcome_ptr": 0.0,
770
+ "predicted_margin": 0.08863633481483887,
771
+ "proposal_oracle_success": 0.0,
772
+ "proposal_oracle_utility": 0.18325337767601013,
773
+ "selected_candidate_index": 8.0,
774
+ "selected_success": 0.0,
775
+ "selected_success_gain_over_base": 0.0,
776
+ "selected_utility": 0.5839591026306152,
777
+ "selected_utility_gain_over_base": 0.0,
778
+ "selector_regret": 0.0,
779
+ "success_selector_gap": 0.0,
780
+ "success_support_gap": 1.0,
781
+ "support_gap": 1.8167466223239899,
782
+ "tau": 0.3473188188503786
783
+ },
784
+ "1522272686/train1": {
785
+ "base_success": 0.0,
786
+ "base_utility": 0.5839591026306152,
787
+ "coverage": 0.0,
788
+ "execute_generated": 0.0,
789
+ "fallback_rate": 1.0,
790
+ "hidden_chart_oracle_success": 1.0,
791
+ "hidden_chart_oracle_utility": 2.0,
792
+ "outcome_ptr": 0.0,
793
+ "predicted_margin": 0.09154681464635614,
794
+ "proposal_oracle_success": 0.0,
795
+ "proposal_oracle_utility": 0.07014383375644684,
796
+ "selected_candidate_index": 7.0,
797
+ "selected_success": 0.0,
798
+ "selected_success_gain_over_base": 0.0,
799
+ "selected_utility": 0.5839591026306152,
800
+ "selected_utility_gain_over_base": 0.0,
801
+ "selector_regret": 0.0,
802
+ "success_selector_gap": 0.0,
803
+ "success_support_gap": 1.0,
804
+ "support_gap": 1.9298561662435532,
805
+ "tau": 0.3473188188503786
806
+ },
807
+ "1522272686/train2": {
808
+ "base_success": 0.0,
809
+ "base_utility": 0.5839591026306152,
810
+ "coverage": 0.0,
811
+ "execute_generated": 0.0,
812
+ "fallback_rate": 1.0,
813
+ "hidden_chart_oracle_success": 1.0,
814
+ "hidden_chart_oracle_utility": 2.0,
815
+ "outcome_ptr": 0.0,
816
+ "predicted_margin": 0.08595256684637552,
817
+ "proposal_oracle_success": 0.0,
818
+ "proposal_oracle_utility": 0.14937306940555573,
819
+ "selected_candidate_index": 8.0,
820
+ "selected_success": 0.0,
821
+ "selected_success_gain_over_base": 0.0,
822
+ "selected_utility": 0.5839591026306152,
823
+ "selected_utility_gain_over_base": 0.0,
824
+ "selector_regret": 0.0,
825
+ "success_selector_gap": 0.0,
826
+ "success_support_gap": 1.0,
827
+ "support_gap": 1.8506269305944443,
828
+ "tau": 0.3473188188503786
829
+ },
830
+ "1549324805/train0": {
831
+ "base_success": 0.0,
832
+ "base_utility": 0.10451483726501465,
833
+ "coverage": 1.0,
834
+ "execute_generated": 1.0,
835
+ "fallback_rate": 0.0,
836
+ "hidden_chart_oracle_success": 0.0,
837
+ "hidden_chart_oracle_utility": 0.2974560856819153,
838
+ "outcome_ptr": 0.0,
839
+ "predicted_margin": 0.4534903350362056,
840
+ "proposal_oracle_success": 0.0,
841
+ "proposal_oracle_utility": 0.060871422290802,
842
+ "selected_candidate_index": 0.0,
843
+ "selected_success": 0.0,
844
+ "selected_success_gain_over_base": 0.0,
845
+ "selected_utility": 0.0260902252048254,
846
+ "selected_utility_gain_over_base": -0.07842461206018925,
847
+ "selector_regret": 0.0347811970859766,
848
+ "success_selector_gap": 0.0,
849
+ "success_support_gap": 0.0,
850
+ "support_gap": 0.23658466339111328,
851
+ "tau": 0.3473188188503786
852
+ },
853
+ "1549324805/train1": {
854
+ "base_success": 0.0,
855
+ "base_utility": 0.10451483726501465,
856
+ "coverage": 0.0,
857
+ "execute_generated": 0.0,
858
+ "fallback_rate": 1.0,
859
+ "hidden_chart_oracle_success": 0.0,
860
+ "hidden_chart_oracle_utility": 0.2974560856819153,
861
+ "outcome_ptr": 1.0,
862
+ "predicted_margin": 0.1768234215359843,
863
+ "proposal_oracle_success": 0.0,
864
+ "proposal_oracle_utility": 0.19892151653766632,
865
+ "selected_candidate_index": 2.0,
866
+ "selected_success": 0.0,
867
+ "selected_success_gain_over_base": 0.0,
868
+ "selected_utility": 0.10451483726501465,
869
+ "selected_utility_gain_over_base": 0.0,
870
+ "selector_regret": 0.09440667927265167,
871
+ "success_selector_gap": 0.0,
872
+ "success_support_gap": 0.0,
873
+ "support_gap": 0.09853456914424896,
874
+ "tau": 0.3473188188503786
875
+ },
876
+ "1549324805/train2": {
877
+ "base_success": 0.0,
878
+ "base_utility": 0.10451483726501465,
879
+ "coverage": 1.0,
880
+ "execute_generated": 1.0,
881
+ "fallback_rate": 0.0,
882
+ "hidden_chart_oracle_success": 0.0,
883
+ "hidden_chart_oracle_utility": 0.2974560856819153,
884
+ "outcome_ptr": 1.0,
885
+ "predicted_margin": 0.5340001574841613,
886
+ "proposal_oracle_success": 0.0,
887
+ "proposal_oracle_utility": 0.11310184001922607,
888
+ "selected_candidate_index": 3.0,
889
+ "selected_success": 0.0,
890
+ "selected_success_gain_over_base": 0.0,
891
+ "selected_utility": 0.02059849165380001,
892
+ "selected_utility_gain_over_base": -0.08391634561121464,
893
+ "selector_regret": 0.09250334836542606,
894
+ "success_selector_gap": 0.0,
895
+ "success_support_gap": 0.0,
896
+ "support_gap": 0.1843542456626892,
897
+ "tau": 0.3473188188503786
898
+ },
899
+ "1578990937/train0": {
900
+ "base_success": 0.0,
901
+ "base_utility": 0.82413649559021,
902
+ "coverage": 0.0,
903
+ "execute_generated": 0.0,
904
+ "fallback_rate": 1.0,
905
+ "hidden_chart_oracle_success": 1.0,
906
+ "hidden_chart_oracle_utility": 2.0,
907
+ "outcome_ptr": 1.0,
908
+ "predicted_margin": 0.2255580985362039,
909
+ "proposal_oracle_success": 1.0,
910
+ "proposal_oracle_utility": 2.0,
911
+ "selected_candidate_index": 15.0,
912
+ "selected_success": 0.0,
913
+ "selected_success_gain_over_base": 0.0,
914
+ "selected_utility": 0.82413649559021,
915
+ "selected_utility_gain_over_base": 0.0,
916
+ "selector_regret": 1.17586350440979,
917
+ "success_selector_gap": 1.0,
918
+ "success_support_gap": 0.0,
919
+ "support_gap": 0.0,
920
+ "tau": 0.3473188188503786
921
+ },
922
+ "1578990937/train1": {
923
+ "base_success": 0.0,
924
+ "base_utility": 0.82413649559021,
925
+ "coverage": 0.0,
926
+ "execute_generated": 0.0,
927
+ "fallback_rate": 1.0,
928
+ "hidden_chart_oracle_success": 1.0,
929
+ "hidden_chart_oracle_utility": 2.0,
930
+ "outcome_ptr": 1.0,
931
+ "predicted_margin": 0.21684013213485268,
932
+ "proposal_oracle_success": 1.0,
933
+ "proposal_oracle_utility": 2.0,
934
+ "selected_candidate_index": 12.0,
935
+ "selected_success": 0.0,
936
+ "selected_success_gain_over_base": 0.0,
937
+ "selected_utility": 0.82413649559021,
938
+ "selected_utility_gain_over_base": 0.0,
939
+ "selector_regret": 1.17586350440979,
940
+ "success_selector_gap": 1.0,
941
+ "success_support_gap": 0.0,
942
+ "support_gap": 0.0,
943
+ "tau": 0.3473188188503786
944
+ },
945
+ "1578990937/train2": {
946
+ "base_success": 0.0,
947
+ "base_utility": 0.82413649559021,
948
+ "coverage": 0.0,
949
+ "execute_generated": 0.0,
950
+ "fallback_rate": 1.0,
951
+ "hidden_chart_oracle_success": 1.0,
952
+ "hidden_chart_oracle_utility": 2.0,
953
+ "outcome_ptr": 1.0,
954
+ "predicted_margin": 0.2250806530775492,
955
+ "proposal_oracle_success": 1.0,
956
+ "proposal_oracle_utility": 2.0,
957
+ "selected_candidate_index": 0.0,
958
+ "selected_success": 0.0,
959
+ "selected_success_gain_over_base": 0.0,
960
+ "selected_utility": 0.82413649559021,
961
+ "selected_utility_gain_over_base": 0.0,
962
+ "selector_regret": 1.17586350440979,
963
+ "success_selector_gap": 1.0,
964
+ "success_support_gap": 0.0,
965
+ "support_gap": 0.0,
966
+ "tau": 0.3473188188503786
967
+ },
968
+ "1747178030/train0": {
969
+ "base_success": 1.0,
970
+ "base_utility": 2.0,
971
+ "coverage": 0.0,
972
+ "execute_generated": 0.0,
973
+ "fallback_rate": 1.0,
974
+ "hidden_chart_oracle_success": 1.0,
975
+ "hidden_chart_oracle_utility": 2.0,
976
+ "outcome_ptr": 0.0,
977
+ "predicted_margin": 0.3352371313866231,
978
+ "proposal_oracle_success": 1.0,
979
+ "proposal_oracle_utility": 2.0,
980
+ "selected_candidate_index": 5.0,
981
+ "selected_success": 1.0,
982
+ "selected_success_gain_over_base": 0.0,
983
+ "selected_utility": 2.0,
984
+ "selected_utility_gain_over_base": 0.0,
985
+ "selector_regret": 0.0,
986
+ "success_selector_gap": 0.0,
987
+ "success_support_gap": 0.0,
988
+ "support_gap": 0.0,
989
+ "tau": 0.3473188188503786
990
+ },
991
+ "1747178030/train1": {
992
+ "base_success": 1.0,
993
+ "base_utility": 2.0,
994
+ "coverage": 0.0,
995
+ "execute_generated": 0.0,
996
+ "fallback_rate": 1.0,
997
+ "hidden_chart_oracle_success": 1.0,
998
+ "hidden_chart_oracle_utility": 2.0,
999
+ "outcome_ptr": 0.0,
1000
+ "predicted_margin": 0.33370414261502274,
1001
+ "proposal_oracle_success": 1.0,
1002
+ "proposal_oracle_utility": 2.0,
1003
+ "selected_candidate_index": 9.0,
1004
+ "selected_success": 1.0,
1005
+ "selected_success_gain_over_base": 0.0,
1006
+ "selected_utility": 2.0,
1007
+ "selected_utility_gain_over_base": 0.0,
1008
+ "selector_regret": 0.0,
1009
+ "success_selector_gap": 0.0,
1010
+ "success_support_gap": 0.0,
1011
+ "support_gap": 0.0,
1012
+ "tau": 0.3473188188503786
1013
+ },
1014
+ "1747178030/train2": {
1015
+ "base_success": 1.0,
1016
+ "base_utility": 2.0,
1017
+ "coverage": 0.0,
1018
+ "execute_generated": 0.0,
1019
+ "fallback_rate": 1.0,
1020
+ "hidden_chart_oracle_success": 1.0,
1021
+ "hidden_chart_oracle_utility": 2.0,
1022
+ "outcome_ptr": 0.0,
1023
+ "predicted_margin": 0.33681729037438607,
1024
+ "proposal_oracle_success": 1.0,
1025
+ "proposal_oracle_utility": 2.0,
1026
+ "selected_candidate_index": 0.0,
1027
+ "selected_success": 1.0,
1028
+ "selected_success_gain_over_base": 0.0,
1029
+ "selected_utility": 2.0,
1030
+ "selected_utility_gain_over_base": 0.0,
1031
+ "selector_regret": 0.0,
1032
+ "success_selector_gap": 0.0,
1033
+ "success_support_gap": 0.0,
1034
+ "support_gap": 0.0,
1035
+ "tau": 0.3473188188503786
1036
+ },
1037
+ "1790188787/train0": {
1038
+ "base_success": 0.0,
1039
+ "base_utility": 0.12476746737957001,
1040
+ "coverage": 1.0,
1041
+ "execute_generated": 1.0,
1042
+ "fallback_rate": 0.0,
1043
+ "hidden_chart_oracle_success": 1.0,
1044
+ "hidden_chart_oracle_utility": 2.0,
1045
+ "outcome_ptr": 1.0,
1046
+ "predicted_margin": 0.7256808008161852,
1047
+ "proposal_oracle_success": 1.0,
1048
+ "proposal_oracle_utility": 2.0,
1049
+ "selected_candidate_index": 9.0,
1050
+ "selected_success": 0.0,
1051
+ "selected_success_gain_over_base": 0.0,
1052
+ "selected_utility": 0.14965972304344177,
1053
+ "selected_utility_gain_over_base": 0.024892255663871765,
1054
+ "selector_regret": 1.8503402769565582,
1055
+ "success_selector_gap": 1.0,
1056
+ "success_support_gap": 0.0,
1057
+ "support_gap": 0.0,
1058
+ "tau": 0.3473188188503786
1059
+ },
1060
+ "1790188787/train1": {
1061
+ "base_success": 0.0,
1062
+ "base_utility": 0.12476746737957001,
1063
+ "coverage": 1.0,
1064
+ "execute_generated": 1.0,
1065
+ "fallback_rate": 0.0,
1066
+ "hidden_chart_oracle_success": 1.0,
1067
+ "hidden_chart_oracle_utility": 2.0,
1068
+ "outcome_ptr": 1.0,
1069
+ "predicted_margin": 0.7407334225672018,
1070
+ "proposal_oracle_success": 1.0,
1071
+ "proposal_oracle_utility": 2.0,
1072
+ "selected_candidate_index": 6.0,
1073
+ "selected_success": 0.0,
1074
+ "selected_success_gain_over_base": 0.0,
1075
+ "selected_utility": 0.13741976022720337,
1076
+ "selected_utility_gain_over_base": 0.012652292847633362,
1077
+ "selector_regret": 1.8625802397727966,
1078
+ "success_selector_gap": 1.0,
1079
+ "success_support_gap": 0.0,
1080
+ "support_gap": 0.0,
1081
+ "tau": 0.3473188188503786
1082
+ },
1083
+ "1790188787/train2": {
1084
+ "base_success": 0.0,
1085
+ "base_utility": 0.12476746737957001,
1086
+ "coverage": 1.0,
1087
+ "execute_generated": 1.0,
1088
+ "fallback_rate": 0.0,
1089
+ "hidden_chart_oracle_success": 1.0,
1090
+ "hidden_chart_oracle_utility": 2.0,
1091
+ "outcome_ptr": 1.0,
1092
+ "predicted_margin": 0.7188844098969164,
1093
+ "proposal_oracle_success": 1.0,
1094
+ "proposal_oracle_utility": 2.0,
1095
+ "selected_candidate_index": 0.0,
1096
+ "selected_success": 1.0,
1097
+ "selected_success_gain_over_base": 1.0,
1098
+ "selected_utility": 2.0,
1099
+ "selected_utility_gain_over_base": 1.87523253262043,
1100
+ "selector_regret": 0.0,
1101
+ "success_selector_gap": 0.0,
1102
+ "success_support_gap": 0.0,
1103
+ "support_gap": 0.0,
1104
+ "tau": 0.3473188188503786
1105
+ },
1106
+ "1822210471/train0": {
1107
+ "base_success": 0.0,
1108
+ "base_utility": 0.07191530615091324,
1109
+ "coverage": 0.0,
1110
+ "execute_generated": 0.0,
1111
+ "fallback_rate": 1.0,
1112
+ "hidden_chart_oracle_success": 1.0,
1113
+ "hidden_chart_oracle_utility": 2.0,
1114
+ "outcome_ptr": 1.0,
1115
+ "predicted_margin": 0.10685249697353241,
1116
+ "proposal_oracle_success": 1.0,
1117
+ "proposal_oracle_utility": 2.0,
1118
+ "selected_candidate_index": 8.0,
1119
+ "selected_success": 0.0,
1120
+ "selected_success_gain_over_base": 0.0,
1121
+ "selected_utility": 0.07191530615091324,
1122
+ "selected_utility_gain_over_base": 0.0,
1123
+ "selector_regret": 1.9280846938490868,
1124
+ "success_selector_gap": 1.0,
1125
+ "success_support_gap": 0.0,
1126
+ "support_gap": 0.0,
1127
+ "tau": 0.3473188188503786
1128
+ },
1129
+ "1822210471/train1": {
1130
+ "base_success": 0.0,
1131
+ "base_utility": 0.07191530615091324,
1132
+ "coverage": 0.0,
1133
+ "execute_generated": 0.0,
1134
+ "fallback_rate": 1.0,
1135
+ "hidden_chart_oracle_success": 1.0,
1136
+ "hidden_chart_oracle_utility": 2.0,
1137
+ "outcome_ptr": 1.0,
1138
+ "predicted_margin": 0.11932515699499706,
1139
+ "proposal_oracle_success": 1.0,
1140
+ "proposal_oracle_utility": 2.0,
1141
+ "selected_candidate_index": 5.0,
1142
+ "selected_success": 0.0,
1143
+ "selected_success_gain_over_base": 0.0,
1144
+ "selected_utility": 0.07191530615091324,
1145
+ "selected_utility_gain_over_base": 0.0,
1146
+ "selector_regret": 1.9280846938490868,
1147
+ "success_selector_gap": 1.0,
1148
+ "success_support_gap": 0.0,
1149
+ "support_gap": 0.0,
1150
+ "tau": 0.3473188188503786
1151
+ },
1152
+ "1822210471/train2": {
1153
+ "base_success": 0.0,
1154
+ "base_utility": 0.07191530615091324,
1155
+ "coverage": 0.0,
1156
+ "execute_generated": 0.0,
1157
+ "fallback_rate": 1.0,
1158
+ "hidden_chart_oracle_success": 1.0,
1159
+ "hidden_chart_oracle_utility": 2.0,
1160
+ "outcome_ptr": 1.0,
1161
+ "predicted_margin": 0.12054968766555163,
1162
+ "proposal_oracle_success": 1.0,
1163
+ "proposal_oracle_utility": 2.0,
1164
+ "selected_candidate_index": 0.0,
1165
+ "selected_success": 0.0,
1166
+ "selected_success_gain_over_base": 0.0,
1167
+ "selected_utility": 0.07191530615091324,
1168
+ "selected_utility_gain_over_base": 0.0,
1169
+ "selector_regret": 1.9280846938490868,
1170
+ "success_selector_gap": 1.0,
1171
+ "success_support_gap": 0.0,
1172
+ "support_gap": 0.0,
1173
+ "tau": 0.3473188188503786
1174
+ },
1175
+ "1845489013/train0": {
1176
+ "base_success": 1.0,
1177
+ "base_utility": 2.0,
1178
+ "coverage": 0.0,
1179
+ "execute_generated": 0.0,
1180
+ "fallback_rate": 1.0,
1181
+ "hidden_chart_oracle_success": 1.0,
1182
+ "hidden_chart_oracle_utility": 2.0,
1183
+ "outcome_ptr": 0.0,
1184
+ "predicted_margin": 0.22822097403519162,
1185
+ "proposal_oracle_success": 1.0,
1186
+ "proposal_oracle_utility": 2.0,
1187
+ "selected_candidate_index": 5.0,
1188
+ "selected_success": 1.0,
1189
+ "selected_success_gain_over_base": 0.0,
1190
+ "selected_utility": 2.0,
1191
+ "selected_utility_gain_over_base": 0.0,
1192
+ "selector_regret": 0.0,
1193
+ "success_selector_gap": 0.0,
1194
+ "success_support_gap": 0.0,
1195
+ "support_gap": 0.0,
1196
+ "tau": 0.3473188188503786
1197
+ },
1198
+ "1845489013/train1": {
1199
+ "base_success": 1.0,
1200
+ "base_utility": 2.0,
1201
+ "coverage": 0.0,
1202
+ "execute_generated": 0.0,
1203
+ "fallback_rate": 1.0,
1204
+ "hidden_chart_oracle_success": 1.0,
1205
+ "hidden_chart_oracle_utility": 2.0,
1206
+ "outcome_ptr": 0.0,
1207
+ "predicted_margin": 0.23290921447763308,
1208
+ "proposal_oracle_success": 1.0,
1209
+ "proposal_oracle_utility": 2.0,
1210
+ "selected_candidate_index": 7.0,
1211
+ "selected_success": 1.0,
1212
+ "selected_success_gain_over_base": 0.0,
1213
+ "selected_utility": 2.0,
1214
+ "selected_utility_gain_over_base": 0.0,
1215
+ "selector_regret": 0.0,
1216
+ "success_selector_gap": 0.0,
1217
+ "success_support_gap": 0.0,
1218
+ "support_gap": 0.0,
1219
+ "tau": 0.3473188188503786
1220
+ },
1221
+ "1845489013/train2": {
1222
+ "base_success": 1.0,
1223
+ "base_utility": 2.0,
1224
+ "coverage": 0.0,
1225
+ "execute_generated": 0.0,
1226
+ "fallback_rate": 1.0,
1227
+ "hidden_chart_oracle_success": 1.0,
1228
+ "hidden_chart_oracle_utility": 2.0,
1229
+ "outcome_ptr": 0.0,
1230
+ "predicted_margin": 0.23237107122905468,
1231
+ "proposal_oracle_success": 1.0,
1232
+ "proposal_oracle_utility": 2.0,
1233
+ "selected_candidate_index": 8.0,
1234
+ "selected_success": 1.0,
1235
+ "selected_success_gain_over_base": 0.0,
1236
+ "selected_utility": 2.0,
1237
+ "selected_utility_gain_over_base": 0.0,
1238
+ "selector_regret": 0.0,
1239
+ "success_selector_gap": 0.0,
1240
+ "success_support_gap": 0.0,
1241
+ "support_gap": 0.0,
1242
+ "tau": 0.3473188188503786
1243
+ },
1244
+ "1999022574/train0": {
1245
+ "base_success": 1.0,
1246
+ "base_utility": 2.0,
1247
+ "coverage": 0.0,
1248
+ "execute_generated": 0.0,
1249
+ "fallback_rate": 1.0,
1250
+ "hidden_chart_oracle_success": 1.0,
1251
+ "hidden_chart_oracle_utility": 2.0,
1252
+ "outcome_ptr": 0.0,
1253
+ "predicted_margin": 0.2830091920285253,
1254
+ "proposal_oracle_success": 1.0,
1255
+ "proposal_oracle_utility": 2.0,
1256
+ "selected_candidate_index": 11.0,
1257
+ "selected_success": 1.0,
1258
+ "selected_success_gain_over_base": 0.0,
1259
+ "selected_utility": 2.0,
1260
+ "selected_utility_gain_over_base": 0.0,
1261
+ "selector_regret": 0.0,
1262
+ "success_selector_gap": 0.0,
1263
+ "success_support_gap": 0.0,
1264
+ "support_gap": 0.0,
1265
+ "tau": 0.3473188188503786
1266
+ },
1267
+ "1999022574/train1": {
1268
+ "base_success": 1.0,
1269
+ "base_utility": 2.0,
1270
+ "coverage": 0.0,
1271
+ "execute_generated": 0.0,
1272
+ "fallback_rate": 1.0,
1273
+ "hidden_chart_oracle_success": 1.0,
1274
+ "hidden_chart_oracle_utility": 2.0,
1275
+ "outcome_ptr": 0.0,
1276
+ "predicted_margin": 0.2762933949412652,
1277
+ "proposal_oracle_success": 1.0,
1278
+ "proposal_oracle_utility": 2.0,
1279
+ "selected_candidate_index": 9.0,
1280
+ "selected_success": 1.0,
1281
+ "selected_success_gain_over_base": 0.0,
1282
+ "selected_utility": 2.0,
1283
+ "selected_utility_gain_over_base": 0.0,
1284
+ "selector_regret": 0.0,
1285
+ "success_selector_gap": 0.0,
1286
+ "success_support_gap": 0.0,
1287
+ "support_gap": 0.0,
1288
+ "tau": 0.3473188188503786
1289
+ },
1290
+ "1999022574/train2": {
1291
+ "base_success": 1.0,
1292
+ "base_utility": 2.0,
1293
+ "coverage": 0.0,
1294
+ "execute_generated": 0.0,
1295
+ "fallback_rate": 1.0,
1296
+ "hidden_chart_oracle_success": 1.0,
1297
+ "hidden_chart_oracle_utility": 2.0,
1298
+ "outcome_ptr": 0.0,
1299
+ "predicted_margin": 0.32995590919912554,
1300
+ "proposal_oracle_success": 1.0,
1301
+ "proposal_oracle_utility": 2.0,
1302
+ "selected_candidate_index": 6.0,
1303
+ "selected_success": 1.0,
1304
+ "selected_success_gain_over_base": 0.0,
1305
+ "selected_utility": 2.0,
1306
+ "selected_utility_gain_over_base": 0.0,
1307
+ "selector_regret": 0.0,
1308
+ "success_selector_gap": 0.0,
1309
+ "success_support_gap": 0.0,
1310
+ "support_gap": 0.0,
1311
+ "tau": 0.3473188188503786
1312
+ },
1313
+ "2012211287/train0": {
1314
+ "base_success": 0.0,
1315
+ "base_utility": 0.7236053347587585,
1316
+ "coverage": 0.0,
1317
+ "execute_generated": 0.0,
1318
+ "fallback_rate": 1.0,
1319
+ "hidden_chart_oracle_success": 1.0,
1320
+ "hidden_chart_oracle_utility": 2.0,
1321
+ "outcome_ptr": 1.0,
1322
+ "predicted_margin": 0.08991720318616597,
1323
+ "proposal_oracle_success": 1.0,
1324
+ "proposal_oracle_utility": 2.0,
1325
+ "selected_candidate_index": 0.0,
1326
+ "selected_success": 0.0,
1327
+ "selected_success_gain_over_base": 0.0,
1328
+ "selected_utility": 0.7236053347587585,
1329
+ "selected_utility_gain_over_base": 0.0,
1330
+ "selector_regret": 1.2763946652412415,
1331
+ "success_selector_gap": 1.0,
1332
+ "success_support_gap": 0.0,
1333
+ "support_gap": 0.0,
1334
+ "tau": 0.3473188188503786
1335
+ },
1336
+ "2012211287/train1": {
1337
+ "base_success": 0.0,
1338
+ "base_utility": 0.7236053347587585,
1339
+ "coverage": 0.0,
1340
+ "execute_generated": 0.0,
1341
+ "fallback_rate": 1.0,
1342
+ "hidden_chart_oracle_success": 1.0,
1343
+ "hidden_chart_oracle_utility": 2.0,
1344
+ "outcome_ptr": 1.0,
1345
+ "predicted_margin": 0.1018745030033766,
1346
+ "proposal_oracle_success": 1.0,
1347
+ "proposal_oracle_utility": 2.0,
1348
+ "selected_candidate_index": 5.0,
1349
+ "selected_success": 0.0,
1350
+ "selected_success_gain_over_base": 0.0,
1351
+ "selected_utility": 0.7236053347587585,
1352
+ "selected_utility_gain_over_base": 0.0,
1353
+ "selector_regret": 1.2763946652412415,
1354
+ "success_selector_gap": 1.0,
1355
+ "success_support_gap": 0.0,
1356
+ "support_gap": 0.0,
1357
+ "tau": 0.3473188188503786
1358
+ },
1359
+ "2012211287/train2": {
1360
+ "base_success": 0.0,
1361
+ "base_utility": 0.7236053347587585,
1362
+ "coverage": 0.0,
1363
+ "execute_generated": 0.0,
1364
+ "fallback_rate": 1.0,
1365
+ "hidden_chart_oracle_success": 1.0,
1366
+ "hidden_chart_oracle_utility": 2.0,
1367
+ "outcome_ptr": 1.0,
1368
+ "predicted_margin": 0.09743460718887578,
1369
+ "proposal_oracle_success": 1.0,
1370
+ "proposal_oracle_utility": 2.0,
1371
+ "selected_candidate_index": 1.0,
1372
+ "selected_success": 0.0,
1373
+ "selected_success_gain_over_base": 0.0,
1374
+ "selected_utility": 0.7236053347587585,
1375
+ "selected_utility_gain_over_base": 0.0,
1376
+ "selector_regret": 1.2763946652412415,
1377
+ "success_selector_gap": 1.0,
1378
+ "success_support_gap": 0.0,
1379
+ "support_gap": 0.0,
1380
+ "tau": 0.3473188188503786
1381
+ },
1382
+ "2038594191/train0": {
1383
+ "base_success": 1.0,
1384
+ "base_utility": 2.0,
1385
+ "coverage": 1.0,
1386
+ "execute_generated": 1.0,
1387
+ "fallback_rate": 0.0,
1388
+ "hidden_chart_oracle_success": 1.0,
1389
+ "hidden_chart_oracle_utility": 2.0,
1390
+ "outcome_ptr": 0.0,
1391
+ "predicted_margin": 0.41980384640853585,
1392
+ "proposal_oracle_success": 1.0,
1393
+ "proposal_oracle_utility": 2.0,
1394
+ "selected_candidate_index": 3.0,
1395
+ "selected_success": 0.0,
1396
+ "selected_success_gain_over_base": -1.0,
1397
+ "selected_utility": 0.05387979745864868,
1398
+ "selected_utility_gain_over_base": -1.9461202025413513,
1399
+ "selector_regret": 1.9461202025413513,
1400
+ "success_selector_gap": 1.0,
1401
+ "success_support_gap": 0.0,
1402
+ "support_gap": 0.0,
1403
+ "tau": 0.3473188188503786
1404
+ },
1405
+ "2038594191/train1": {
1406
+ "base_success": 1.0,
1407
+ "base_utility": 2.0,
1408
+ "coverage": 1.0,
1409
+ "execute_generated": 1.0,
1410
+ "fallback_rate": 0.0,
1411
+ "hidden_chart_oracle_success": 1.0,
1412
+ "hidden_chart_oracle_utility": 2.0,
1413
+ "outcome_ptr": 0.0,
1414
+ "predicted_margin": 0.39227093434151067,
1415
+ "proposal_oracle_success": 1.0,
1416
+ "proposal_oracle_utility": 2.0,
1417
+ "selected_candidate_index": 0.0,
1418
+ "selected_success": 0.0,
1419
+ "selected_success_gain_over_base": -1.0,
1420
+ "selected_utility": 0.023384133353829384,
1421
+ "selected_utility_gain_over_base": -1.9766158666461706,
1422
+ "selector_regret": 1.9766158666461706,
1423
+ "success_selector_gap": 1.0,
1424
+ "success_support_gap": 0.0,
1425
+ "support_gap": 0.0,
1426
+ "tau": 0.3473188188503786
1427
+ },
1428
+ "2038594191/train2": {
1429
+ "base_success": 1.0,
1430
+ "base_utility": 2.0,
1431
+ "coverage": 1.0,
1432
+ "execute_generated": 1.0,
1433
+ "fallback_rate": 0.0,
1434
+ "hidden_chart_oracle_success": 1.0,
1435
+ "hidden_chart_oracle_utility": 2.0,
1436
+ "outcome_ptr": 0.0,
1437
+ "predicted_margin": 0.41980384640853585,
1438
+ "proposal_oracle_success": 1.0,
1439
+ "proposal_oracle_utility": 2.0,
1440
+ "selected_candidate_index": 1.0,
1441
+ "selected_success": 1.0,
1442
+ "selected_success_gain_over_base": 0.0,
1443
+ "selected_utility": 2.0,
1444
+ "selected_utility_gain_over_base": 0.0,
1445
+ "selector_regret": 0.0,
1446
+ "success_selector_gap": 0.0,
1447
+ "success_support_gap": 0.0,
1448
+ "support_gap": 0.0,
1449
+ "tau": 0.3473188188503786
1450
+ },
1451
+ "2054899126/train0": {
1452
+ "base_success": 1.0,
1453
+ "base_utility": 2.0,
1454
+ "coverage": 1.0,
1455
+ "execute_generated": 1.0,
1456
+ "fallback_rate": 0.0,
1457
+ "hidden_chart_oracle_success": 1.0,
1458
+ "hidden_chart_oracle_utility": 2.0,
1459
+ "outcome_ptr": 0.0,
1460
+ "predicted_margin": 0.48682363081170954,
1461
+ "proposal_oracle_success": 0.5,
1462
+ "proposal_oracle_utility": 1.077283151447773,
1463
+ "selected_candidate_index": 0.0,
1464
+ "selected_success": 0.5,
1465
+ "selected_success_gain_over_base": -0.5,
1466
+ "selected_utility": 1.052448581904173,
1467
+ "selected_utility_gain_over_base": -0.9475514180958271,
1468
+ "selector_regret": 0.024834569543600082,
1469
+ "success_selector_gap": 0.0,
1470
+ "success_support_gap": 0.5,
1471
+ "support_gap": 0.922716848552227,
1472
+ "tau": 0.3473188188503786
1473
+ },
1474
+ "2054899126/train1": {
1475
+ "base_success": 1.0,
1476
+ "base_utility": 2.0,
1477
+ "coverage": 1.0,
1478
+ "execute_generated": 1.0,
1479
+ "fallback_rate": 0.0,
1480
+ "hidden_chart_oracle_success": 1.0,
1481
+ "hidden_chart_oracle_utility": 2.0,
1482
+ "outcome_ptr": 0.0,
1483
+ "predicted_margin": 0.4867067691213679,
1484
+ "proposal_oracle_success": 1.0,
1485
+ "proposal_oracle_utility": 2.0,
1486
+ "selected_candidate_index": 0.0,
1487
+ "selected_success": 0.5,
1488
+ "selected_success_gain_over_base": -0.5,
1489
+ "selected_utility": 1.0664558708667755,
1490
+ "selected_utility_gain_over_base": -0.9335441291332245,
1491
+ "selector_regret": 0.9335441291332245,
1492
+ "success_selector_gap": 0.5,
1493
+ "success_support_gap": 0.0,
1494
+ "support_gap": 0.0,
1495
+ "tau": 0.3473188188503786
1496
+ },
1497
+ "2054899126/train2": {
1498
+ "base_success": 1.0,
1499
+ "base_utility": 2.0,
1500
+ "coverage": 0.5,
1501
+ "execute_generated": 0.5,
1502
+ "fallback_rate": 0.5,
1503
+ "hidden_chart_oracle_success": 1.0,
1504
+ "hidden_chart_oracle_utility": 2.0,
1505
+ "outcome_ptr": 0.0,
1506
+ "predicted_margin": 0.4238219648796263,
1507
+ "proposal_oracle_success": 1.0,
1508
+ "proposal_oracle_utility": 2.0,
1509
+ "selected_candidate_index": 0.5,
1510
+ "selected_success": 1.0,
1511
+ "selected_success_gain_over_base": 0.0,
1512
+ "selected_utility": 2.0,
1513
+ "selected_utility_gain_over_base": 0.0,
1514
+ "selector_regret": 0.0,
1515
+ "success_selector_gap": 0.0,
1516
+ "success_support_gap": 0.0,
1517
+ "support_gap": 0.0,
1518
+ "tau": 0.3473188188503786
1519
+ },
1520
+ "2061635888/train0": {
1521
+ "base_success": 0.0,
1522
+ "base_utility": 0.05225050449371338,
1523
+ "coverage": 0.0,
1524
+ "execute_generated": 0.0,
1525
+ "fallback_rate": 1.0,
1526
+ "hidden_chart_oracle_success": 1.0,
1527
+ "hidden_chart_oracle_utility": 2.0,
1528
+ "outcome_ptr": 1.0,
1529
+ "predicted_margin": 0.16138880459399857,
1530
+ "proposal_oracle_success": 0.0,
1531
+ "proposal_oracle_utility": 0.09799930453300476,
1532
+ "selected_candidate_index": 5.0,
1533
+ "selected_success": 0.0,
1534
+ "selected_success_gain_over_base": 0.0,
1535
+ "selected_utility": 0.05225050449371338,
1536
+ "selected_utility_gain_over_base": 0.0,
1537
+ "selector_regret": 0.04574880003929138,
1538
+ "success_selector_gap": 0.0,
1539
+ "success_support_gap": 1.0,
1540
+ "support_gap": 1.9020006954669952,
1541
+ "tau": 0.3473188188503786
1542
+ },
1543
+ "2061635888/train1": {
1544
+ "base_success": 0.0,
1545
+ "base_utility": 0.05225050449371338,
1546
+ "coverage": 0.0,
1547
+ "execute_generated": 0.0,
1548
+ "fallback_rate": 1.0,
1549
+ "hidden_chart_oracle_success": 1.0,
1550
+ "hidden_chart_oracle_utility": 2.0,
1551
+ "outcome_ptr": 1.0,
1552
+ "predicted_margin": 0.12788381852052721,
1553
+ "proposal_oracle_success": 0.0,
1554
+ "proposal_oracle_utility": 0.07580336928367615,
1555
+ "selected_candidate_index": 6.0,
1556
+ "selected_success": 0.0,
1557
+ "selected_success_gain_over_base": 0.0,
1558
+ "selected_utility": 0.05225050449371338,
1559
+ "selected_utility_gain_over_base": 0.0,
1560
+ "selector_regret": 0.02355286478996277,
1561
+ "success_selector_gap": 0.0,
1562
+ "success_support_gap": 1.0,
1563
+ "support_gap": 1.9241966307163239,
1564
+ "tau": 0.3473188188503786
1565
+ },
1566
+ "2061635888/train2": {
1567
+ "base_success": 0.0,
1568
+ "base_utility": 0.05225050449371338,
1569
+ "coverage": 0.0,
1570
+ "execute_generated": 0.0,
1571
+ "fallback_rate": 1.0,
1572
+ "hidden_chart_oracle_success": 1.0,
1573
+ "hidden_chart_oracle_utility": 2.0,
1574
+ "outcome_ptr": 1.0,
1575
+ "predicted_margin": 0.13169501516378973,
1576
+ "proposal_oracle_success": 0.0,
1577
+ "proposal_oracle_utility": 0.11105132102966309,
1578
+ "selected_candidate_index": 4.0,
1579
+ "selected_success": 0.0,
1580
+ "selected_success_gain_over_base": 0.0,
1581
+ "selected_utility": 0.05225050449371338,
1582
+ "selected_utility_gain_over_base": 0.0,
1583
+ "selector_regret": 0.05880081653594971,
1584
+ "success_selector_gap": 0.0,
1585
+ "success_support_gap": 1.0,
1586
+ "support_gap": 1.888948678970337,
1587
+ "tau": 0.3473188188503786
1588
+ },
1589
+ "211245842/train0": {
1590
+ "base_success": 0.0,
1591
+ "base_utility": 0.19090841710567474,
1592
+ "coverage": 1.0,
1593
+ "execute_generated": 1.0,
1594
+ "fallback_rate": 0.0,
1595
+ "hidden_chart_oracle_success": 1.0,
1596
+ "hidden_chart_oracle_utility": 2.0,
1597
+ "outcome_ptr": 1.0,
1598
+ "predicted_margin": 0.36446911237871227,
1599
+ "proposal_oracle_success": 1.0,
1600
+ "proposal_oracle_utility": 2.0,
1601
+ "selected_candidate_index": 7.0,
1602
+ "selected_success": 0.0,
1603
+ "selected_success_gain_over_base": 0.0,
1604
+ "selected_utility": 0.11833827197551727,
1605
+ "selected_utility_gain_over_base": -0.07257014513015747,
1606
+ "selector_regret": 1.8816617280244827,
1607
+ "success_selector_gap": 1.0,
1608
+ "success_support_gap": 0.0,
1609
+ "support_gap": 0.0,
1610
+ "tau": 0.3473188188503786
1611
+ },
1612
+ "211245842/train1": {
1613
+ "base_success": 0.0,
1614
+ "base_utility": 0.19090841710567474,
1615
+ "coverage": 1.0,
1616
+ "execute_generated": 1.0,
1617
+ "fallback_rate": 0.0,
1618
+ "hidden_chart_oracle_success": 1.0,
1619
+ "hidden_chart_oracle_utility": 2.0,
1620
+ "outcome_ptr": 1.0,
1621
+ "predicted_margin": 0.3818735579959133,
1622
+ "proposal_oracle_success": 0.0,
1623
+ "proposal_oracle_utility": 0.21295717358589172,
1624
+ "selected_candidate_index": 13.0,
1625
+ "selected_success": 0.0,
1626
+ "selected_success_gain_over_base": 0.0,
1627
+ "selected_utility": 0.17156320810317993,
1628
+ "selected_utility_gain_over_base": -0.019345209002494812,
1629
+ "selector_regret": 0.04139396548271179,
1630
+ "success_selector_gap": 0.0,
1631
+ "success_support_gap": 1.0,
1632
+ "support_gap": 1.7870428264141083,
1633
+ "tau": 0.3473188188503786
1634
+ },
1635
+ "211245842/train2": {
1636
+ "base_success": 0.0,
1637
+ "base_utility": 0.19090841710567474,
1638
+ "coverage": 1.0,
1639
+ "execute_generated": 1.0,
1640
+ "fallback_rate": 0.0,
1641
+ "hidden_chart_oracle_success": 1.0,
1642
+ "hidden_chart_oracle_utility": 2.0,
1643
+ "outcome_ptr": 1.0,
1644
+ "predicted_margin": 0.3818735579959133,
1645
+ "proposal_oracle_success": 0.0,
1646
+ "proposal_oracle_utility": 0.23388276994228363,
1647
+ "selected_candidate_index": 0.0,
1648
+ "selected_success": 0.0,
1649
+ "selected_success_gain_over_base": 0.0,
1650
+ "selected_utility": 0.17691531777381897,
1651
+ "selected_utility_gain_over_base": -0.013993099331855774,
1652
+ "selector_regret": 0.05696745216846466,
1653
+ "success_selector_gap": 0.0,
1654
+ "success_support_gap": 1.0,
1655
+ "support_gap": 1.7661172300577164,
1656
+ "tau": 0.3473188188503786
1657
+ },
1658
+ "278008524/train0": {
1659
+ "base_success": 0.0,
1660
+ "base_utility": 0.042734989896416664,
1661
+ "coverage": 1.0,
1662
+ "execute_generated": 1.0,
1663
+ "fallback_rate": 0.0,
1664
+ "hidden_chart_oracle_success": 0.5,
1665
+ "hidden_chart_oracle_utility": 1.1139086410403252,
1666
+ "outcome_ptr": 1.0,
1667
+ "predicted_margin": 0.4154810514015457,
1668
+ "proposal_oracle_success": 0.5,
1669
+ "proposal_oracle_utility": 1.0733563229441643,
1670
+ "selected_candidate_index": 3.0,
1671
+ "selected_success": 0.5,
1672
+ "selected_success_gain_over_base": 0.5,
1673
+ "selected_utility": 1.0485825426876545,
1674
+ "selected_utility_gain_over_base": 1.0058475527912378,
1675
+ "selector_regret": 0.02477378025650978,
1676
+ "success_selector_gap": 0.0,
1677
+ "success_support_gap": 0.0,
1678
+ "support_gap": 0.04055231809616089,
1679
+ "tau": 0.3473188188503786
1680
+ },
1681
+ "278008524/train1": {
1682
+ "base_success": 0.0,
1683
+ "base_utility": 0.042734989896416664,
1684
+ "coverage": 1.0,
1685
+ "execute_generated": 1.0,
1686
+ "fallback_rate": 0.0,
1687
+ "hidden_chart_oracle_success": 0.5,
1688
+ "hidden_chart_oracle_utility": 1.1139086410403252,
1689
+ "outcome_ptr": 1.0,
1690
+ "predicted_margin": 0.431062905506948,
1691
+ "proposal_oracle_success": 0.5,
1692
+ "proposal_oracle_utility": 1.0376745946705341,
1693
+ "selected_candidate_index": 1.0,
1694
+ "selected_success": 0.0,
1695
+ "selected_success_gain_over_base": 0.0,
1696
+ "selected_utility": 0.1001356840133667,
1697
+ "selected_utility_gain_over_base": 0.057400694116950035,
1698
+ "selector_regret": 0.9375389106571674,
1699
+ "success_selector_gap": 0.5,
1700
+ "success_support_gap": 0.0,
1701
+ "support_gap": 0.07623404636979103,
1702
+ "tau": 0.3473188188503786
1703
+ },
1704
+ "278008524/train2": {
1705
+ "base_success": 0.0,
1706
+ "base_utility": 0.042734989896416664,
1707
+ "coverage": 0.5,
1708
+ "execute_generated": 0.5,
1709
+ "fallback_rate": 0.5,
1710
+ "hidden_chart_oracle_success": 0.5,
1711
+ "hidden_chart_oracle_utility": 1.1139086410403252,
1712
+ "outcome_ptr": 1.0,
1713
+ "predicted_margin": 0.39125281310385696,
1714
+ "proposal_oracle_success": 0.5,
1715
+ "proposal_oracle_utility": 1.1359489113092422,
1716
+ "selected_candidate_index": 11.5,
1717
+ "selected_success": 0.0,
1718
+ "selected_success_gain_over_base": 0.0,
1719
+ "selected_utility": 0.052897630259394646,
1720
+ "selected_utility_gain_over_base": 0.010162640362977982,
1721
+ "selector_regret": 1.0830512810498476,
1722
+ "success_selector_gap": 0.5,
1723
+ "success_support_gap": 0.0,
1724
+ "support_gap": 0.0,
1725
+ "tau": 0.3473188188503786
1726
+ },
1727
+ "307068879/train0": {
1728
+ "base_success": 0.0,
1729
+ "base_utility": 0.6000905632972717,
1730
+ "coverage": 0.0,
1731
+ "execute_generated": 0.0,
1732
+ "fallback_rate": 1.0,
1733
+ "hidden_chart_oracle_success": 1.0,
1734
+ "hidden_chart_oracle_utility": 2.0,
1735
+ "outcome_ptr": 0.0,
1736
+ "predicted_margin": 0.2797080480367002,
1737
+ "proposal_oracle_success": 0.0,
1738
+ "proposal_oracle_utility": 0.2115575522184372,
1739
+ "selected_candidate_index": 2.0,
1740
+ "selected_success": 0.0,
1741
+ "selected_success_gain_over_base": 0.0,
1742
+ "selected_utility": 0.6000905632972717,
1743
+ "selected_utility_gain_over_base": 0.0,
1744
+ "selector_regret": 0.0,
1745
+ "success_selector_gap": 0.0,
1746
+ "success_support_gap": 1.0,
1747
+ "support_gap": 1.7884424477815628,
1748
+ "tau": 0.3473188188503786
1749
+ },
1750
+ "307068879/train1": {
1751
+ "base_success": 0.0,
1752
+ "base_utility": 0.6000905632972717,
1753
+ "coverage": 0.0,
1754
+ "execute_generated": 0.0,
1755
+ "fallback_rate": 1.0,
1756
+ "hidden_chart_oracle_success": 1.0,
1757
+ "hidden_chart_oracle_utility": 2.0,
1758
+ "outcome_ptr": 0.0,
1759
+ "predicted_margin": 0.24778606153341362,
1760
+ "proposal_oracle_success": 0.0,
1761
+ "proposal_oracle_utility": 0.5876073241233826,
1762
+ "selected_candidate_index": 12.0,
1763
+ "selected_success": 0.0,
1764
+ "selected_success_gain_over_base": 0.0,
1765
+ "selected_utility": 0.6000905632972717,
1766
+ "selected_utility_gain_over_base": 0.0,
1767
+ "selector_regret": 0.0,
1768
+ "success_selector_gap": 0.0,
1769
+ "success_support_gap": 1.0,
1770
+ "support_gap": 1.4123926758766174,
1771
+ "tau": 0.3473188188503786
1772
+ },
1773
+ "307068879/train2": {
1774
+ "base_success": 0.0,
1775
+ "base_utility": 0.6000905632972717,
1776
+ "coverage": 0.0,
1777
+ "execute_generated": 0.0,
1778
+ "fallback_rate": 1.0,
1779
+ "hidden_chart_oracle_success": 1.0,
1780
+ "hidden_chart_oracle_utility": 2.0,
1781
+ "outcome_ptr": 0.0,
1782
+ "predicted_margin": 0.29487688188584205,
1783
+ "proposal_oracle_success": 0.0,
1784
+ "proposal_oracle_utility": 0.2302134484052658,
1785
+ "selected_candidate_index": 0.0,
1786
+ "selected_success": 0.0,
1787
+ "selected_success_gain_over_base": 0.0,
1788
+ "selected_utility": 0.6000905632972717,
1789
+ "selected_utility_gain_over_base": 0.0,
1790
+ "selector_regret": 0.0,
1791
+ "success_selector_gap": 0.0,
1792
+ "success_support_gap": 1.0,
1793
+ "support_gap": 1.7697865515947342,
1794
+ "tau": 0.3473188188503786
1795
+ },
1796
+ "412102563/train0": {
1797
+ "base_success": 1.0,
1798
+ "base_utility": 2.0,
1799
+ "coverage": 0.0,
1800
+ "execute_generated": 0.0,
1801
+ "fallback_rate": 1.0,
1802
+ "hidden_chart_oracle_success": 1.0,
1803
+ "hidden_chart_oracle_utility": 2.0,
1804
+ "outcome_ptr": 0.0,
1805
+ "predicted_margin": 0.20585303592937382,
1806
+ "proposal_oracle_success": 1.0,
1807
+ "proposal_oracle_utility": 2.0,
1808
+ "selected_candidate_index": 6.0,
1809
+ "selected_success": 1.0,
1810
+ "selected_success_gain_over_base": 0.0,
1811
+ "selected_utility": 2.0,
1812
+ "selected_utility_gain_over_base": 0.0,
1813
+ "selector_regret": 0.0,
1814
+ "success_selector_gap": 0.0,
1815
+ "success_support_gap": 0.0,
1816
+ "support_gap": 0.0,
1817
+ "tau": 0.3473188188503786
1818
+ },
1819
+ "412102563/train1": {
1820
+ "base_success": 1.0,
1821
+ "base_utility": 2.0,
1822
+ "coverage": 0.0,
1823
+ "execute_generated": 0.0,
1824
+ "fallback_rate": 1.0,
1825
+ "hidden_chart_oracle_success": 1.0,
1826
+ "hidden_chart_oracle_utility": 2.0,
1827
+ "outcome_ptr": 0.0,
1828
+ "predicted_margin": 0.20493799872831076,
1829
+ "proposal_oracle_success": 1.0,
1830
+ "proposal_oracle_utility": 2.0,
1831
+ "selected_candidate_index": 10.0,
1832
+ "selected_success": 1.0,
1833
+ "selected_success_gain_over_base": 0.0,
1834
+ "selected_utility": 2.0,
1835
+ "selected_utility_gain_over_base": 0.0,
1836
+ "selector_regret": 0.0,
1837
+ "success_selector_gap": 0.0,
1838
+ "success_support_gap": 0.0,
1839
+ "support_gap": 0.0,
1840
+ "tau": 0.3473188188503786
1841
+ },
1842
+ "412102563/train2": {
1843
+ "base_success": 1.0,
1844
+ "base_utility": 2.0,
1845
+ "coverage": 0.0,
1846
+ "execute_generated": 0.0,
1847
+ "fallback_rate": 1.0,
1848
+ "hidden_chart_oracle_success": 1.0,
1849
+ "hidden_chart_oracle_utility": 2.0,
1850
+ "outcome_ptr": 0.0,
1851
+ "predicted_margin": 0.18125860724557305,
1852
+ "proposal_oracle_success": 1.0,
1853
+ "proposal_oracle_utility": 2.0,
1854
+ "selected_candidate_index": 3.0,
1855
+ "selected_success": 1.0,
1856
+ "selected_success_gain_over_base": 0.0,
1857
+ "selected_utility": 2.0,
1858
+ "selected_utility_gain_over_base": 0.0,
1859
+ "selector_regret": 0.0,
1860
+ "success_selector_gap": 0.0,
1861
+ "success_support_gap": 0.0,
1862
+ "support_gap": 0.0,
1863
+ "tau": 0.3473188188503786
1864
+ },
1865
+ "440577711/train0": {
1866
+ "base_success": 0.0,
1867
+ "base_utility": 0.036152780055999756,
1868
+ "coverage": 0.0,
1869
+ "execute_generated": 0.0,
1870
+ "fallback_rate": 1.0,
1871
+ "hidden_chart_oracle_success": 1.0,
1872
+ "hidden_chart_oracle_utility": 2.0,
1873
+ "outcome_ptr": 1.0,
1874
+ "predicted_margin": 0.1872847414980603,
1875
+ "proposal_oracle_success": 1.0,
1876
+ "proposal_oracle_utility": 2.0,
1877
+ "selected_candidate_index": 3.0,
1878
+ "selected_success": 0.0,
1879
+ "selected_success_gain_over_base": 0.0,
1880
+ "selected_utility": 0.036152780055999756,
1881
+ "selected_utility_gain_over_base": 0.0,
1882
+ "selector_regret": 1.9638472199440002,
1883
+ "success_selector_gap": 1.0,
1884
+ "success_support_gap": 0.0,
1885
+ "support_gap": 0.0,
1886
+ "tau": 0.3473188188503786
1887
+ },
1888
+ "440577711/train1": {
1889
+ "base_success": 0.0,
1890
+ "base_utility": 0.036152780055999756,
1891
+ "coverage": 0.0,
1892
+ "execute_generated": 0.0,
1893
+ "fallback_rate": 1.0,
1894
+ "hidden_chart_oracle_success": 1.0,
1895
+ "hidden_chart_oracle_utility": 2.0,
1896
+ "outcome_ptr": 1.0,
1897
+ "predicted_margin": 0.1872847414980603,
1898
+ "proposal_oracle_success": 1.0,
1899
+ "proposal_oracle_utility": 2.0,
1900
+ "selected_candidate_index": 5.0,
1901
+ "selected_success": 0.0,
1902
+ "selected_success_gain_over_base": 0.0,
1903
+ "selected_utility": 0.036152780055999756,
1904
+ "selected_utility_gain_over_base": 0.0,
1905
+ "selector_regret": 1.9638472199440002,
1906
+ "success_selector_gap": 1.0,
1907
+ "success_support_gap": 0.0,
1908
+ "support_gap": 0.0,
1909
+ "tau": 0.3473188188503786
1910
+ },
1911
+ "440577711/train2": {
1912
+ "base_success": 0.0,
1913
+ "base_utility": 0.036152780055999756,
1914
+ "coverage": 0.0,
1915
+ "execute_generated": 0.0,
1916
+ "fallback_rate": 1.0,
1917
+ "hidden_chart_oracle_success": 1.0,
1918
+ "hidden_chart_oracle_utility": 2.0,
1919
+ "outcome_ptr": 1.0,
1920
+ "predicted_margin": 0.1872847414980603,
1921
+ "proposal_oracle_success": 1.0,
1922
+ "proposal_oracle_utility": 2.0,
1923
+ "selected_candidate_index": 5.0,
1924
+ "selected_success": 0.0,
1925
+ "selected_success_gain_over_base": 0.0,
1926
+ "selected_utility": 0.036152780055999756,
1927
+ "selected_utility_gain_over_base": 0.0,
1928
+ "selector_regret": 1.9638472199440002,
1929
+ "success_selector_gap": 1.0,
1930
+ "success_support_gap": 0.0,
1931
+ "support_gap": 0.0,
1932
+ "tau": 0.3473188188503786
1933
+ },
1934
+ "441046872/train0": {
1935
+ "base_success": 0.0,
1936
+ "base_utility": 0.6999551653862,
1937
+ "coverage": 0.0,
1938
+ "execute_generated": 0.0,
1939
+ "fallback_rate": 1.0,
1940
+ "hidden_chart_oracle_success": 1.0,
1941
+ "hidden_chart_oracle_utility": 2.0,
1942
+ "outcome_ptr": 1.0,
1943
+ "predicted_margin": 0.20172868695503152,
1944
+ "proposal_oracle_success": 1.0,
1945
+ "proposal_oracle_utility": 2.0,
1946
+ "selected_candidate_index": 10.0,
1947
+ "selected_success": 0.0,
1948
+ "selected_success_gain_over_base": 0.0,
1949
+ "selected_utility": 0.6999551653862,
1950
+ "selected_utility_gain_over_base": 0.0,
1951
+ "selector_regret": 1.3000448346138,
1952
+ "success_selector_gap": 1.0,
1953
+ "success_support_gap": 0.0,
1954
+ "support_gap": 0.0,
1955
+ "tau": 0.3473188188503786
1956
+ },
1957
+ "441046872/train1": {
1958
+ "base_success": 0.0,
1959
+ "base_utility": 0.6999551653862,
1960
+ "coverage": 0.0,
1961
+ "execute_generated": 0.0,
1962
+ "fallback_rate": 1.0,
1963
+ "hidden_chart_oracle_success": 1.0,
1964
+ "hidden_chart_oracle_utility": 2.0,
1965
+ "outcome_ptr": 1.0,
1966
+ "predicted_margin": 0.24443771108596596,
1967
+ "proposal_oracle_success": 0.0,
1968
+ "proposal_oracle_utility": 0.7233588695526123,
1969
+ "selected_candidate_index": 3.0,
1970
+ "selected_success": 0.0,
1971
+ "selected_success_gain_over_base": 0.0,
1972
+ "selected_utility": 0.6999551653862,
1973
+ "selected_utility_gain_over_base": 0.0,
1974
+ "selector_regret": 0.023403704166412354,
1975
+ "success_selector_gap": 0.0,
1976
+ "success_support_gap": 1.0,
1977
+ "support_gap": 1.2766411304473877,
1978
+ "tau": 0.3473188188503786
1979
+ },
1980
+ "441046872/train2": {
1981
+ "base_success": 0.0,
1982
+ "base_utility": 0.6999551653862,
1983
+ "coverage": 0.0,
1984
+ "execute_generated": 0.0,
1985
+ "fallback_rate": 1.0,
1986
+ "hidden_chart_oracle_success": 1.0,
1987
+ "hidden_chart_oracle_utility": 2.0,
1988
+ "outcome_ptr": 1.0,
1989
+ "predicted_margin": 0.19710526958727811,
1990
+ "proposal_oracle_success": 0.0,
1991
+ "proposal_oracle_utility": 0.7191348671913147,
1992
+ "selected_candidate_index": 0.0,
1993
+ "selected_success": 0.0,
1994
+ "selected_success_gain_over_base": 0.0,
1995
+ "selected_utility": 0.6999551653862,
1996
+ "selected_utility_gain_over_base": 0.0,
1997
+ "selector_regret": 0.019179701805114746,
1998
+ "success_selector_gap": 0.0,
1999
+ "success_support_gap": 1.0,
2000
+ "support_gap": 1.2808651328086853,
2001
+ "tau": 0.3473188188503786
2002
+ },
2003
+ "509430392/train0": {
2004
+ "base_success": 0.0,
2005
+ "base_utility": 2.2351741790771484e-07,
2006
+ "coverage": 1.0,
2007
+ "execute_generated": 1.0,
2008
+ "fallback_rate": 0.0,
2009
+ "hidden_chart_oracle_success": 1.0,
2010
+ "hidden_chart_oracle_utility": 2.0,
2011
+ "outcome_ptr": 1.0,
2012
+ "predicted_margin": 0.39582323274190295,
2013
+ "proposal_oracle_success": 1.0,
2014
+ "proposal_oracle_utility": 2.0,
2015
+ "selected_candidate_index": 14.0,
2016
+ "selected_success": 0.0,
2017
+ "selected_success_gain_over_base": 0.0,
2018
+ "selected_utility": 0.12097799777984619,
2019
+ "selected_utility_gain_over_base": 0.12097777426242828,
2020
+ "selector_regret": 1.8790220022201538,
2021
+ "success_selector_gap": 1.0,
2022
+ "success_support_gap": 0.0,
2023
+ "support_gap": 0.0,
2024
+ "tau": 0.3473188188503786
2025
+ },
2026
+ "509430392/train1": {
2027
+ "base_success": 0.0,
2028
+ "base_utility": 2.2351741790771484e-07,
2029
+ "coverage": 1.0,
2030
+ "execute_generated": 1.0,
2031
+ "fallback_rate": 0.0,
2032
+ "hidden_chart_oracle_success": 1.0,
2033
+ "hidden_chart_oracle_utility": 2.0,
2034
+ "outcome_ptr": 1.0,
2035
+ "predicted_margin": 0.3836105423742797,
2036
+ "proposal_oracle_success": 1.0,
2037
+ "proposal_oracle_utility": 2.0,
2038
+ "selected_candidate_index": 2.0,
2039
+ "selected_success": 1.0,
2040
+ "selected_success_gain_over_base": 1.0,
2041
+ "selected_utility": 2.0,
2042
+ "selected_utility_gain_over_base": 1.999999776482582,
2043
+ "selector_regret": 0.0,
2044
+ "success_selector_gap": 0.0,
2045
+ "success_support_gap": 0.0,
2046
+ "support_gap": 0.0,
2047
+ "tau": 0.3473188188503786
2048
+ },
2049
+ "509430392/train2": {
2050
+ "base_success": 0.0,
2051
+ "base_utility": 2.2351741790771484e-07,
2052
+ "coverage": 1.0,
2053
+ "execute_generated": 1.0,
2054
+ "fallback_rate": 0.0,
2055
+ "hidden_chart_oracle_success": 1.0,
2056
+ "hidden_chart_oracle_utility": 2.0,
2057
+ "outcome_ptr": 1.0,
2058
+ "predicted_margin": 0.3982631871077599,
2059
+ "proposal_oracle_success": 0.0,
2060
+ "proposal_oracle_utility": 5.424022674560547e-06,
2061
+ "selected_candidate_index": 1.0,
2062
+ "selected_success": 0.0,
2063
+ "selected_success_gain_over_base": 0.0,
2064
+ "selected_utility": 2.682209014892578e-07,
2065
+ "selected_utility_gain_over_base": 4.470348358154297e-08,
2066
+ "selector_regret": 5.155801773071289e-06,
2067
+ "success_selector_gap": 0.0,
2068
+ "success_support_gap": 1.0,
2069
+ "support_gap": 1.9999945759773254,
2070
+ "tau": 0.3473188188503786
2071
+ },
2072
+ "585904130/train0": {
2073
+ "base_success": 0.0,
2074
+ "base_utility": 0.18171554803848267,
2075
+ "coverage": 0.0,
2076
+ "execute_generated": 0.0,
2077
+ "fallback_rate": 1.0,
2078
+ "hidden_chart_oracle_success": 1.0,
2079
+ "hidden_chart_oracle_utility": 2.0,
2080
+ "outcome_ptr": 1.0,
2081
+ "predicted_margin": 0.26430287849596273,
2082
+ "proposal_oracle_success": 1.0,
2083
+ "proposal_oracle_utility": 2.0,
2084
+ "selected_candidate_index": 14.0,
2085
+ "selected_success": 0.0,
2086
+ "selected_success_gain_over_base": 0.0,
2087
+ "selected_utility": 0.18171554803848267,
2088
+ "selected_utility_gain_over_base": 0.0,
2089
+ "selector_regret": 1.8182844519615173,
2090
+ "success_selector_gap": 1.0,
2091
+ "success_support_gap": 0.0,
2092
+ "support_gap": 0.0,
2093
+ "tau": 0.3473188188503786
2094
+ },
2095
+ "585904130/train1": {
2096
+ "base_success": 0.0,
2097
+ "base_utility": 0.18171554803848267,
2098
+ "coverage": 0.0,
2099
+ "execute_generated": 0.0,
2100
+ "fallback_rate": 1.0,
2101
+ "hidden_chart_oracle_success": 1.0,
2102
+ "hidden_chart_oracle_utility": 2.0,
2103
+ "outcome_ptr": 1.0,
2104
+ "predicted_margin": 0.23332367249842212,
2105
+ "proposal_oracle_success": 1.0,
2106
+ "proposal_oracle_utility": 2.0,
2107
+ "selected_candidate_index": 7.0,
2108
+ "selected_success": 0.0,
2109
+ "selected_success_gain_over_base": 0.0,
2110
+ "selected_utility": 0.18171554803848267,
2111
+ "selected_utility_gain_over_base": 0.0,
2112
+ "selector_regret": 1.8182844519615173,
2113
+ "success_selector_gap": 1.0,
2114
+ "success_support_gap": 0.0,
2115
+ "support_gap": 0.0,
2116
+ "tau": 0.3473188188503786
2117
+ },
2118
+ "585904130/train2": {
2119
+ "base_success": 0.0,
2120
+ "base_utility": 0.18171554803848267,
2121
+ "coverage": 0.0,
2122
+ "execute_generated": 0.0,
2123
+ "fallback_rate": 1.0,
2124
+ "hidden_chart_oracle_success": 1.0,
2125
+ "hidden_chart_oracle_utility": 2.0,
2126
+ "outcome_ptr": 1.0,
2127
+ "predicted_margin": 0.2531734715082984,
2128
+ "proposal_oracle_success": 1.0,
2129
+ "proposal_oracle_utility": 2.0,
2130
+ "selected_candidate_index": 4.0,
2131
+ "selected_success": 0.0,
2132
+ "selected_success_gain_over_base": 0.0,
2133
+ "selected_utility": 0.18171554803848267,
2134
+ "selected_utility_gain_over_base": 0.0,
2135
+ "selector_regret": 1.8182844519615173,
2136
+ "success_selector_gap": 1.0,
2137
+ "success_support_gap": 0.0,
2138
+ "support_gap": 0.0,
2139
+ "tau": 0.3473188188503786
2140
+ },
2141
+ "608016290/train0": {
2142
+ "base_success": 0.0,
2143
+ "base_utility": 0.725701630115509,
2144
+ "coverage": 0.0,
2145
+ "execute_generated": 0.0,
2146
+ "fallback_rate": 1.0,
2147
+ "hidden_chart_oracle_success": 1.0,
2148
+ "hidden_chart_oracle_utility": 2.0,
2149
+ "outcome_ptr": 1.0,
2150
+ "predicted_margin": 0.18560038716365618,
2151
+ "proposal_oracle_success": 1.0,
2152
+ "proposal_oracle_utility": 2.0,
2153
+ "selected_candidate_index": 3.0,
2154
+ "selected_success": 0.0,
2155
+ "selected_success_gain_over_base": 0.0,
2156
+ "selected_utility": 0.725701630115509,
2157
+ "selected_utility_gain_over_base": 0.0,
2158
+ "selector_regret": 1.274298369884491,
2159
+ "success_selector_gap": 1.0,
2160
+ "success_support_gap": 0.0,
2161
+ "support_gap": 0.0,
2162
+ "tau": 0.3473188188503786
2163
+ },
2164
+ "608016290/train1": {
2165
+ "base_success": 0.0,
2166
+ "base_utility": 0.725701630115509,
2167
+ "coverage": 0.0,
2168
+ "execute_generated": 0.0,
2169
+ "fallback_rate": 1.0,
2170
+ "hidden_chart_oracle_success": 1.0,
2171
+ "hidden_chart_oracle_utility": 2.0,
2172
+ "outcome_ptr": 1.0,
2173
+ "predicted_margin": 0.2379598925580104,
2174
+ "proposal_oracle_success": 1.0,
2175
+ "proposal_oracle_utility": 2.0,
2176
+ "selected_candidate_index": 8.0,
2177
+ "selected_success": 0.0,
2178
+ "selected_success_gain_over_base": 0.0,
2179
+ "selected_utility": 0.725701630115509,
2180
+ "selected_utility_gain_over_base": 0.0,
2181
+ "selector_regret": 1.274298369884491,
2182
+ "success_selector_gap": 1.0,
2183
+ "success_support_gap": 0.0,
2184
+ "support_gap": 0.0,
2185
+ "tau": 0.3473188188503786
2186
+ },
2187
+ "608016290/train2": {
2188
+ "base_success": 0.0,
2189
+ "base_utility": 0.725701630115509,
2190
+ "coverage": 0.0,
2191
+ "execute_generated": 0.0,
2192
+ "fallback_rate": 1.0,
2193
+ "hidden_chart_oracle_success": 1.0,
2194
+ "hidden_chart_oracle_utility": 2.0,
2195
+ "outcome_ptr": 0.0,
2196
+ "predicted_margin": 0.1808325968136315,
2197
+ "proposal_oracle_success": 0.0,
2198
+ "proposal_oracle_utility": 0.7117276191711426,
2199
+ "selected_candidate_index": 15.0,
2200
+ "selected_success": 0.0,
2201
+ "selected_success_gain_over_base": 0.0,
2202
+ "selected_utility": 0.725701630115509,
2203
+ "selected_utility_gain_over_base": 0.0,
2204
+ "selector_regret": 0.0,
2205
+ "success_selector_gap": 0.0,
2206
+ "success_support_gap": 1.0,
2207
+ "support_gap": 1.2882723808288574,
2208
+ "tau": 0.3473188188503786
2209
+ },
2210
+ "626486813/train0": {
2211
+ "base_success": 0.0,
2212
+ "base_utility": 0.00509834272088483,
2213
+ "coverage": 0.5,
2214
+ "execute_generated": 0.5,
2215
+ "fallback_rate": 0.5,
2216
+ "hidden_chart_oracle_success": 0.0,
2217
+ "hidden_chart_oracle_utility": 0.16558238118886948,
2218
+ "outcome_ptr": 1.0,
2219
+ "predicted_margin": 0.41319658430741574,
2220
+ "proposal_oracle_success": 0.0,
2221
+ "proposal_oracle_utility": 0.04469547560438514,
2222
+ "selected_candidate_index": 0.0,
2223
+ "selected_success": 0.0,
2224
+ "selected_success_gain_over_base": 0.0,
2225
+ "selected_utility": 0.005386819248087704,
2226
+ "selected_utility_gain_over_base": 0.0002884765272028744,
2227
+ "selector_regret": 0.03930865635629743,
2228
+ "success_selector_gap": 0.0,
2229
+ "success_support_gap": 0.0,
2230
+ "support_gap": 0.12161866808310151,
2231
+ "tau": 0.3473188188503786
2232
+ },
2233
+ "626486813/train1": {
2234
+ "base_success": 0.0,
2235
+ "base_utility": 0.00509834272088483,
2236
+ "coverage": 0.5,
2237
+ "execute_generated": 0.5,
2238
+ "fallback_rate": 0.5,
2239
+ "hidden_chart_oracle_success": 0.0,
2240
+ "hidden_chart_oracle_utility": 0.16558238118886948,
2241
+ "outcome_ptr": 1.0,
2242
+ "predicted_margin": 0.46817656253554407,
2243
+ "proposal_oracle_success": 0.0,
2244
+ "proposal_oracle_utility": 0.035827874671667814,
2245
+ "selected_candidate_index": 0.5,
2246
+ "selected_success": 0.0,
2247
+ "selected_success_gain_over_base": 0.0,
2248
+ "selected_utility": 0.00787139916792512,
2249
+ "selected_utility_gain_over_base": 0.0027730564470402896,
2250
+ "selector_regret": 0.027956475503742695,
2251
+ "success_selector_gap": 0.0,
2252
+ "success_support_gap": 0.0,
2253
+ "support_gap": 0.12975450651720166,
2254
+ "tau": 0.3473188188503786
2255
+ },
2256
+ "626486813/train2": {
2257
+ "base_success": 0.0,
2258
+ "base_utility": 0.00509834272088483,
2259
+ "coverage": 0.5,
2260
+ "execute_generated": 0.5,
2261
+ "fallback_rate": 0.5,
2262
+ "hidden_chart_oracle_success": 0.0,
2263
+ "hidden_chart_oracle_utility": 0.16558238118886948,
2264
+ "outcome_ptr": 1.0,
2265
+ "predicted_margin": 0.3841254242297905,
2266
+ "proposal_oracle_success": 0.0,
2267
+ "proposal_oracle_utility": 0.04489925503730774,
2268
+ "selected_candidate_index": 1.0,
2269
+ "selected_success": 0.0,
2270
+ "selected_success_gain_over_base": 0.0,
2271
+ "selected_utility": 0.015375842805951834,
2272
+ "selected_utility_gain_over_base": 0.010277500085067004,
2273
+ "selector_regret": 0.029523412231355906,
2274
+ "success_selector_gap": 0.0,
2275
+ "success_support_gap": 0.0,
2276
+ "support_gap": 0.12068312615156174,
2277
+ "tau": 0.3473188188503786
2278
+ },
2279
+ "636680137/train0": {
2280
+ "base_success": 0.0,
2281
+ "base_utility": 0.08313504606485367,
2282
+ "coverage": 1.0,
2283
+ "execute_generated": 1.0,
2284
+ "fallback_rate": 0.0,
2285
+ "hidden_chart_oracle_success": 0.0,
2286
+ "hidden_chart_oracle_utility": 0.17056432366371155,
2287
+ "outcome_ptr": 1.0,
2288
+ "predicted_margin": 0.5371194860392053,
2289
+ "proposal_oracle_success": 0.0,
2290
+ "proposal_oracle_utility": 0.08826130628585815,
2291
+ "selected_candidate_index": 0.0,
2292
+ "selected_success": 0.0,
2293
+ "selected_success_gain_over_base": 0.0,
2294
+ "selected_utility": 0.056092143058776855,
2295
+ "selected_utility_gain_over_base": -0.027042903006076813,
2296
+ "selector_regret": 0.0321691632270813,
2297
+ "success_selector_gap": 0.0,
2298
+ "success_support_gap": 0.0,
2299
+ "support_gap": 0.0823030173778534,
2300
+ "tau": 0.3473188188503786
2301
+ },
2302
+ "636680137/train1": {
2303
+ "base_success": 0.0,
2304
+ "base_utility": 0.08313504606485367,
2305
+ "coverage": 0.0,
2306
+ "execute_generated": 0.0,
2307
+ "fallback_rate": 1.0,
2308
+ "hidden_chart_oracle_success": 0.0,
2309
+ "hidden_chart_oracle_utility": 0.17056432366371155,
2310
+ "outcome_ptr": 0.0,
2311
+ "predicted_margin": 0.2496823681236075,
2312
+ "proposal_oracle_success": 0.0,
2313
+ "proposal_oracle_utility": 0.06741118431091309,
2314
+ "selected_candidate_index": 0.0,
2315
+ "selected_success": 0.0,
2316
+ "selected_success_gain_over_base": 0.0,
2317
+ "selected_utility": 0.08313504606485367,
2318
+ "selected_utility_gain_over_base": 0.0,
2319
+ "selector_regret": 0.0,
2320
+ "success_selector_gap": 0.0,
2321
+ "success_support_gap": 0.0,
2322
+ "support_gap": 0.10315313935279846,
2323
+ "tau": 0.3473188188503786
2324
+ },
2325
+ "636680137/train2": {
2326
+ "base_success": 0.0,
2327
+ "base_utility": 0.08313504606485367,
2328
+ "coverage": 0.0,
2329
+ "execute_generated": 0.0,
2330
+ "fallback_rate": 1.0,
2331
+ "hidden_chart_oracle_success": 0.0,
2332
+ "hidden_chart_oracle_utility": 0.17056432366371155,
2333
+ "outcome_ptr": 0.0,
2334
+ "predicted_margin": 0.2912910141265051,
2335
+ "proposal_oracle_success": 0.0,
2336
+ "proposal_oracle_utility": 0.0756516233086586,
2337
+ "selected_candidate_index": 1.0,
2338
+ "selected_success": 0.0,
2339
+ "selected_success_gain_over_base": 0.0,
2340
+ "selected_utility": 0.08313504606485367,
2341
+ "selected_utility_gain_over_base": 0.0,
2342
+ "selector_regret": 0.0,
2343
+ "success_selector_gap": 0.0,
2344
+ "success_support_gap": 0.0,
2345
+ "support_gap": 0.09491270035505295,
2346
+ "tau": 0.3473188188503786
2347
+ },
2348
+ "655884605/train0": {
2349
+ "base_success": 0.0,
2350
+ "base_utility": 0.16300667822360992,
2351
+ "coverage": 0.0,
2352
+ "execute_generated": 0.0,
2353
+ "fallback_rate": 1.0,
2354
+ "hidden_chart_oracle_success": 1.0,
2355
+ "hidden_chart_oracle_utility": 2.0,
2356
+ "outcome_ptr": 0.0,
2357
+ "predicted_margin": 0.30368062376907057,
2358
+ "proposal_oracle_success": 0.0,
2359
+ "proposal_oracle_utility": 0.13732868432998657,
2360
+ "selected_candidate_index": 1.0,
2361
+ "selected_success": 0.0,
2362
+ "selected_success_gain_over_base": 0.0,
2363
+ "selected_utility": 0.16300667822360992,
2364
+ "selected_utility_gain_over_base": 0.0,
2365
+ "selector_regret": 0.0,
2366
+ "success_selector_gap": 0.0,
2367
+ "success_support_gap": 1.0,
2368
+ "support_gap": 1.8626713156700134,
2369
+ "tau": 0.3473188188503786
2370
+ },
2371
+ "655884605/train1": {
2372
+ "base_success": 0.0,
2373
+ "base_utility": 0.16300667822360992,
2374
+ "coverage": 0.0,
2375
+ "execute_generated": 0.0,
2376
+ "fallback_rate": 1.0,
2377
+ "hidden_chart_oracle_success": 1.0,
2378
+ "hidden_chart_oracle_utility": 2.0,
2379
+ "outcome_ptr": 1.0,
2380
+ "predicted_margin": 0.2713166038796752,
2381
+ "proposal_oracle_success": 0.0,
2382
+ "proposal_oracle_utility": 0.8128758072853088,
2383
+ "selected_candidate_index": 2.0,
2384
+ "selected_success": 0.0,
2385
+ "selected_success_gain_over_base": 0.0,
2386
+ "selected_utility": 0.16300667822360992,
2387
+ "selected_utility_gain_over_base": 0.0,
2388
+ "selector_regret": 0.6498691290616989,
2389
+ "success_selector_gap": 0.0,
2390
+ "success_support_gap": 1.0,
2391
+ "support_gap": 1.1871241927146912,
2392
+ "tau": 0.3473188188503786
2393
+ },
2394
+ "655884605/train2": {
2395
+ "base_success": 0.0,
2396
+ "base_utility": 0.16300667822360992,
2397
+ "coverage": 0.0,
2398
+ "execute_generated": 0.0,
2399
+ "fallback_rate": 1.0,
2400
+ "hidden_chart_oracle_success": 1.0,
2401
+ "hidden_chart_oracle_utility": 2.0,
2402
+ "outcome_ptr": 1.0,
2403
+ "predicted_margin": 0.19620741382881715,
2404
+ "proposal_oracle_success": 0.0,
2405
+ "proposal_oracle_utility": 0.21447330713272095,
2406
+ "selected_candidate_index": 8.0,
2407
+ "selected_success": 0.0,
2408
+ "selected_success_gain_over_base": 0.0,
2409
+ "selected_utility": 0.16300667822360992,
2410
+ "selected_utility_gain_over_base": 0.0,
2411
+ "selector_regret": 0.05146662890911102,
2412
+ "success_selector_gap": 0.0,
2413
+ "success_support_gap": 1.0,
2414
+ "support_gap": 1.785526692867279,
2415
+ "tau": 0.3473188188503786
2416
+ },
2417
+ "672328262/train0": {
2418
+ "base_success": 1.0,
2419
+ "base_utility": 2.0,
2420
+ "coverage": 0.0,
2421
+ "execute_generated": 0.0,
2422
+ "fallback_rate": 1.0,
2423
+ "hidden_chart_oracle_success": 1.0,
2424
+ "hidden_chart_oracle_utility": 2.0,
2425
+ "outcome_ptr": 0.0,
2426
+ "predicted_margin": 0.042802515015988585,
2427
+ "proposal_oracle_success": 1.0,
2428
+ "proposal_oracle_utility": 2.0,
2429
+ "selected_candidate_index": 8.0,
2430
+ "selected_success": 1.0,
2431
+ "selected_success_gain_over_base": 0.0,
2432
+ "selected_utility": 2.0,
2433
+ "selected_utility_gain_over_base": 0.0,
2434
+ "selector_regret": 0.0,
2435
+ "success_selector_gap": 0.0,
2436
+ "success_support_gap": 0.0,
2437
+ "support_gap": 0.0,
2438
+ "tau": 0.3473188188503786
2439
+ },
2440
+ "672328262/train1": {
2441
+ "base_success": 1.0,
2442
+ "base_utility": 2.0,
2443
+ "coverage": 0.0,
2444
+ "execute_generated": 0.0,
2445
+ "fallback_rate": 1.0,
2446
+ "hidden_chart_oracle_success": 1.0,
2447
+ "hidden_chart_oracle_utility": 2.0,
2448
+ "outcome_ptr": 0.0,
2449
+ "predicted_margin": 0.04466194101303914,
2450
+ "proposal_oracle_success": 1.0,
2451
+ "proposal_oracle_utility": 2.0,
2452
+ "selected_candidate_index": 1.0,
2453
+ "selected_success": 1.0,
2454
+ "selected_success_gain_over_base": 0.0,
2455
+ "selected_utility": 2.0,
2456
+ "selected_utility_gain_over_base": 0.0,
2457
+ "selector_regret": 0.0,
2458
+ "success_selector_gap": 0.0,
2459
+ "success_support_gap": 0.0,
2460
+ "support_gap": 0.0,
2461
+ "tau": 0.3473188188503786
2462
+ },
2463
+ "672328262/train2": {
2464
+ "base_success": 1.0,
2465
+ "base_utility": 2.0,
2466
+ "coverage": 0.0,
2467
+ "execute_generated": 0.0,
2468
+ "fallback_rate": 1.0,
2469
+ "hidden_chart_oracle_success": 1.0,
2470
+ "hidden_chart_oracle_utility": 2.0,
2471
+ "outcome_ptr": 0.0,
2472
+ "predicted_margin": 0.048270193237023314,
2473
+ "proposal_oracle_success": 1.0,
2474
+ "proposal_oracle_utility": 2.0,
2475
+ "selected_candidate_index": 2.0,
2476
+ "selected_success": 1.0,
2477
+ "selected_success_gain_over_base": 0.0,
2478
+ "selected_utility": 2.0,
2479
+ "selected_utility_gain_over_base": 0.0,
2480
+ "selector_regret": 0.0,
2481
+ "success_selector_gap": 0.0,
2482
+ "success_support_gap": 0.0,
2483
+ "support_gap": 0.0,
2484
+ "tau": 0.3473188188503786
2485
+ },
2486
+ "701667486/train0": {
2487
+ "base_success": 1.0,
2488
+ "base_utility": 2.0,
2489
+ "coverage": 1.0,
2490
+ "execute_generated": 1.0,
2491
+ "fallback_rate": 0.0,
2492
+ "hidden_chart_oracle_success": 1.0,
2493
+ "hidden_chart_oracle_utility": 2.0,
2494
+ "outcome_ptr": 0.0,
2495
+ "predicted_margin": 0.3589247640152091,
2496
+ "proposal_oracle_success": 1.0,
2497
+ "proposal_oracle_utility": 2.0,
2498
+ "selected_candidate_index": 2.0,
2499
+ "selected_success": 1.0,
2500
+ "selected_success_gain_over_base": 0.0,
2501
+ "selected_utility": 2.0,
2502
+ "selected_utility_gain_over_base": 0.0,
2503
+ "selector_regret": 0.0,
2504
+ "success_selector_gap": 0.0,
2505
+ "success_support_gap": 0.0,
2506
+ "support_gap": 0.0,
2507
+ "tau": 0.3473188188503786
2508
+ },
2509
+ "701667486/train1": {
2510
+ "base_success": 1.0,
2511
+ "base_utility": 2.0,
2512
+ "coverage": 1.0,
2513
+ "execute_generated": 1.0,
2514
+ "fallback_rate": 0.0,
2515
+ "hidden_chart_oracle_success": 1.0,
2516
+ "hidden_chart_oracle_utility": 2.0,
2517
+ "outcome_ptr": 0.0,
2518
+ "predicted_margin": 0.38798745623031156,
2519
+ "proposal_oracle_success": 1.0,
2520
+ "proposal_oracle_utility": 2.0,
2521
+ "selected_candidate_index": 10.0,
2522
+ "selected_success": 1.0,
2523
+ "selected_success_gain_over_base": 0.0,
2524
+ "selected_utility": 2.0,
2525
+ "selected_utility_gain_over_base": 0.0,
2526
+ "selector_regret": 0.0,
2527
+ "success_selector_gap": 0.0,
2528
+ "success_support_gap": 0.0,
2529
+ "support_gap": 0.0,
2530
+ "tau": 0.3473188188503786
2531
+ },
2532
+ "701667486/train2": {
2533
+ "base_success": 1.0,
2534
+ "base_utility": 2.0,
2535
+ "coverage": 1.0,
2536
+ "execute_generated": 1.0,
2537
+ "fallback_rate": 0.0,
2538
+ "hidden_chart_oracle_success": 1.0,
2539
+ "hidden_chart_oracle_utility": 2.0,
2540
+ "outcome_ptr": 0.0,
2541
+ "predicted_margin": 0.4021201653187239,
2542
+ "proposal_oracle_success": 1.0,
2543
+ "proposal_oracle_utility": 2.0,
2544
+ "selected_candidate_index": 8.0,
2545
+ "selected_success": 1.0,
2546
+ "selected_success_gain_over_base": 0.0,
2547
+ "selected_utility": 2.0,
2548
+ "selected_utility_gain_over_base": 0.0,
2549
+ "selector_regret": 0.0,
2550
+ "success_selector_gap": 0.0,
2551
+ "success_support_gap": 0.0,
2552
+ "support_gap": 0.0,
2553
+ "tau": 0.3473188188503786
2554
+ },
2555
+ "747260437/train0": {
2556
+ "base_success": 1.0,
2557
+ "base_utility": 2.0,
2558
+ "coverage": 0.0,
2559
+ "execute_generated": 0.0,
2560
+ "fallback_rate": 1.0,
2561
+ "hidden_chart_oracle_success": 0.0,
2562
+ "hidden_chart_oracle_utility": 0.2786632478237152,
2563
+ "outcome_ptr": 0.0,
2564
+ "predicted_margin": 0.28297507531186483,
2565
+ "proposal_oracle_success": 0.0,
2566
+ "proposal_oracle_utility": 0.10839563608169556,
2567
+ "selected_candidate_index": 0.0,
2568
+ "selected_success": 1.0,
2569
+ "selected_success_gain_over_base": 0.0,
2570
+ "selected_utility": 2.0,
2571
+ "selected_utility_gain_over_base": 0.0,
2572
+ "selector_regret": 0.0,
2573
+ "success_selector_gap": 0.0,
2574
+ "success_support_gap": 0.0,
2575
+ "support_gap": 0.17026761174201965,
2576
+ "tau": 0.3473188188503786
2577
+ },
2578
+ "747260437/train1": {
2579
+ "base_success": 1.0,
2580
+ "base_utility": 2.0,
2581
+ "coverage": 0.0,
2582
+ "execute_generated": 0.0,
2583
+ "fallback_rate": 1.0,
2584
+ "hidden_chart_oracle_success": 0.0,
2585
+ "hidden_chart_oracle_utility": 0.2786632478237152,
2586
+ "outcome_ptr": 0.0,
2587
+ "predicted_margin": 0.2525195047503382,
2588
+ "proposal_oracle_success": 0.0,
2589
+ "proposal_oracle_utility": 0.06328952312469482,
2590
+ "selected_candidate_index": 8.0,
2591
+ "selected_success": 1.0,
2592
+ "selected_success_gain_over_base": 0.0,
2593
+ "selected_utility": 2.0,
2594
+ "selected_utility_gain_over_base": 0.0,
2595
+ "selector_regret": 0.0,
2596
+ "success_selector_gap": 0.0,
2597
+ "success_support_gap": 0.0,
2598
+ "support_gap": 0.21537372469902039,
2599
+ "tau": 0.3473188188503786
2600
+ },
2601
+ "747260437/train2": {
2602
+ "base_success": 1.0,
2603
+ "base_utility": 2.0,
2604
+ "coverage": 1.0,
2605
+ "execute_generated": 1.0,
2606
+ "fallback_rate": 0.0,
2607
+ "hidden_chart_oracle_success": 0.0,
2608
+ "hidden_chart_oracle_utility": 0.2786632478237152,
2609
+ "outcome_ptr": 0.0,
2610
+ "predicted_margin": 0.5259115214300562,
2611
+ "proposal_oracle_success": 0.0,
2612
+ "proposal_oracle_utility": 0.12510187923908234,
2613
+ "selected_candidate_index": 1.0,
2614
+ "selected_success": 0.0,
2615
+ "selected_success_gain_over_base": -1.0,
2616
+ "selected_utility": 0.06632735580205917,
2617
+ "selected_utility_gain_over_base": -1.9336726441979408,
2618
+ "selector_regret": 0.05877452343702316,
2619
+ "success_selector_gap": 0.0,
2620
+ "success_support_gap": 0.0,
2621
+ "support_gap": 0.15356136858463287,
2622
+ "tau": 0.3473188188503786
2623
+ },
2624
+ "750448296/train0": {
2625
+ "base_success": 0.0,
2626
+ "base_utility": 0.11759314934412639,
2627
+ "coverage": 0.3333333333333333,
2628
+ "execute_generated": 0.3333333333333333,
2629
+ "fallback_rate": 0.6666666666666666,
2630
+ "hidden_chart_oracle_success": 0.6666666666666666,
2631
+ "hidden_chart_oracle_utility": 1.4079964955647786,
2632
+ "outcome_ptr": 0.6666666666666666,
2633
+ "predicted_margin": 0.4190984327494606,
2634
+ "proposal_oracle_success": 0.3333333333333333,
2635
+ "proposal_oracle_utility": 0.7526603279014429,
2636
+ "selected_candidate_index": 4.0,
2637
+ "selected_success": 0.0,
2638
+ "selected_success_gain_over_base": 0.0,
2639
+ "selected_utility": 0.09317883538703124,
2640
+ "selected_utility_gain_over_base": -0.024414313957095146,
2641
+ "selector_regret": 0.6594814925144116,
2642
+ "success_selector_gap": 0.3333333333333333,
2643
+ "success_support_gap": 0.3333333333333333,
2644
+ "support_gap": 0.6553361676633358,
2645
+ "tau": 0.3473188188503786
2646
+ },
2647
+ "750448296/train1": {
2648
+ "base_success": 0.0,
2649
+ "base_utility": 0.11759314934412639,
2650
+ "coverage": 0.3333333333333333,
2651
+ "execute_generated": 0.3333333333333333,
2652
+ "fallback_rate": 0.6666666666666666,
2653
+ "hidden_chart_oracle_success": 0.6666666666666666,
2654
+ "hidden_chart_oracle_utility": 1.4079964955647786,
2655
+ "outcome_ptr": 1.0,
2656
+ "predicted_margin": 0.33489946089892014,
2657
+ "proposal_oracle_success": 0.6666666666666666,
2658
+ "proposal_oracle_utility": 1.3705032741030057,
2659
+ "selected_candidate_index": 3.0,
2660
+ "selected_success": 0.0,
2661
+ "selected_success_gain_over_base": 0.0,
2662
+ "selected_utility": 0.11618362615505855,
2663
+ "selected_utility_gain_over_base": -0.0014095231890678406,
2664
+ "selector_regret": 1.2543196479479473,
2665
+ "success_selector_gap": 0.6666666666666666,
2666
+ "success_support_gap": 0.0,
2667
+ "support_gap": 0.03749322146177292,
2668
+ "tau": 0.3473188188503786
2669
+ },
2670
+ "750448296/train2": {
2671
+ "base_success": 0.0,
2672
+ "base_utility": 0.11759314934412639,
2673
+ "coverage": 0.3333333333333333,
2674
+ "execute_generated": 0.3333333333333333,
2675
+ "fallback_rate": 0.6666666666666666,
2676
+ "hidden_chart_oracle_success": 0.6666666666666666,
2677
+ "hidden_chart_oracle_utility": 1.4079964955647786,
2678
+ "outcome_ptr": 0.6666666666666666,
2679
+ "predicted_margin": 0.31678157342373975,
2680
+ "proposal_oracle_success": 0.6666666666666666,
2681
+ "proposal_oracle_utility": 1.3504486680030823,
2682
+ "selected_candidate_index": 2.0,
2683
+ "selected_success": 0.0,
2684
+ "selected_success_gain_over_base": 0.0,
2685
+ "selected_utility": 0.1037093202273051,
2686
+ "selected_utility_gain_over_base": -0.013883829116821289,
2687
+ "selector_regret": 1.246739347775777,
2688
+ "success_selector_gap": 0.6666666666666666,
2689
+ "success_support_gap": 0.0,
2690
+ "support_gap": 0.05754782756169637,
2691
+ "tau": 0.3473188188503786
2692
+ },
2693
+ "767747424/train0": {
2694
+ "base_success": 0.25,
2695
+ "base_utility": 0.5731993420049548,
2696
+ "coverage": 0.25,
2697
+ "execute_generated": 0.25,
2698
+ "fallback_rate": 0.75,
2699
+ "hidden_chart_oracle_success": 0.25,
2700
+ "hidden_chart_oracle_utility": 0.6737305596470833,
2701
+ "outcome_ptr": 0.5,
2702
+ "predicted_margin": 0.28067674068264487,
2703
+ "proposal_oracle_success": 0.5,
2704
+ "proposal_oracle_utility": 1.0424291878007352,
2705
+ "selected_candidate_index": 7.0,
2706
+ "selected_success": 0.25,
2707
+ "selected_success_gain_over_base": 0.0,
2708
+ "selected_utility": 0.5386653137393296,
2709
+ "selected_utility_gain_over_base": -0.03453402826562524,
2710
+ "selector_regret": 0.5124941510148346,
2711
+ "success_selector_gap": 0.25,
2712
+ "success_support_gap": 0.0,
2713
+ "support_gap": 0.06562284985557199,
2714
+ "tau": 0.3473188188503786
2715
+ },
2716
+ "767747424/train1": {
2717
+ "base_success": 0.25,
2718
+ "base_utility": 0.5731993420049548,
2719
+ "coverage": 0.25,
2720
+ "execute_generated": 0.25,
2721
+ "fallback_rate": 0.75,
2722
+ "hidden_chart_oracle_success": 0.25,
2723
+ "hidden_chart_oracle_utility": 0.6737305596470833,
2724
+ "outcome_ptr": 0.5,
2725
+ "predicted_margin": 0.28220702596299274,
2726
+ "proposal_oracle_success": 0.5,
2727
+ "proposal_oracle_utility": 1.0629530912265182,
2728
+ "selected_candidate_index": 2.0,
2729
+ "selected_success": 0.25,
2730
+ "selected_success_gain_over_base": 0.0,
2731
+ "selected_utility": 0.5581304961815476,
2732
+ "selected_utility_gain_over_base": -0.015068845823407173,
2733
+ "selector_regret": 0.5076578482985497,
2734
+ "success_selector_gap": 0.25,
2735
+ "success_support_gap": 0.0,
2736
+ "support_gap": 0.045098946429789066,
2737
+ "tau": 0.3473188188503786
2738
+ },
2739
+ "767747424/train2": {
2740
+ "base_success": 0.25,
2741
+ "base_utility": 0.5731993420049548,
2742
+ "coverage": 0.25,
2743
+ "execute_generated": 0.25,
2744
+ "fallback_rate": 0.75,
2745
+ "hidden_chart_oracle_success": 0.25,
2746
+ "hidden_chart_oracle_utility": 0.6737305596470833,
2747
+ "outcome_ptr": 0.5,
2748
+ "predicted_margin": 0.28169149441007413,
2749
+ "proposal_oracle_success": 0.5,
2750
+ "proposal_oracle_utility": 1.0671670949086547,
2751
+ "selected_candidate_index": 7.75,
2752
+ "selected_success": 0.25,
2753
+ "selected_success_gain_over_base": 0.0,
2754
+ "selected_utility": 0.5799880577251315,
2755
+ "selected_utility_gain_over_base": 0.006788715720176697,
2756
+ "selector_regret": 0.48825800977647305,
2757
+ "success_selector_gap": 0.25,
2758
+ "success_support_gap": 0.0,
2759
+ "support_gap": 0.04088494274765253,
2760
+ "tau": 0.3473188188503786
2761
+ }
2762
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/metrics_by_task.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "LiftPegUpright-v1": {
3
+ "base_success": 0.2,
4
+ "base_utility": 0.9498705387115478,
5
+ "coverage": 0.0,
6
+ "execute_generated": 0.0,
7
+ "fallback_rate": 1.0,
8
+ "hidden_chart_oracle_success": 1.0,
9
+ "hidden_chart_oracle_utility": 2.0,
10
+ "outcome_ptr": 0.5333333333333333,
11
+ "predicted_margin": 0.21812074468121995,
12
+ "proposal_oracle_success": 0.6,
13
+ "proposal_oracle_utility": 1.4122399787108104,
14
+ "selected_candidate_index": 5.0,
15
+ "selected_success": 0.2,
16
+ "selected_success_gain_over_base": 0.0,
17
+ "selected_utility": 0.9498705387115478,
18
+ "selected_utility_gain_over_base": 0.0,
19
+ "selector_regret": 0.5146939317385356,
20
+ "success_selector_gap": 0.4,
21
+ "success_support_gap": 0.4,
22
+ "support_gap": 0.5877600212891897,
23
+ "tau": 0.34731881885037846
24
+ },
25
+ "PickCube-v1": {
26
+ "base_success": 0.0,
27
+ "base_utility": 0.5839591026306152,
28
+ "coverage": 0.0,
29
+ "execute_generated": 0.0,
30
+ "fallback_rate": 1.0,
31
+ "hidden_chart_oracle_success": 1.0,
32
+ "hidden_chart_oracle_utility": 2.0,
33
+ "outcome_ptr": 0.0,
34
+ "predicted_margin": 0.08871190543585684,
35
+ "proposal_oracle_success": 0.0,
36
+ "proposal_oracle_utility": 0.13425676027933756,
37
+ "selected_candidate_index": 7.666666666666667,
38
+ "selected_success": 0.0,
39
+ "selected_success_gain_over_base": 0.0,
40
+ "selected_utility": 0.5839591026306152,
41
+ "selected_utility_gain_over_base": 0.0,
42
+ "selector_regret": 0.0,
43
+ "success_selector_gap": 0.0,
44
+ "success_support_gap": 1.0,
45
+ "support_gap": 1.8657432397206624,
46
+ "tau": 0.3473188188503786
47
+ },
48
+ "PullCube-v1": {
49
+ "base_success": 0.36,
50
+ "base_utility": 0.7751704612234608,
51
+ "coverage": 0.38666666666666666,
52
+ "execute_generated": 0.38666666666666666,
53
+ "fallback_rate": 0.6133333333333333,
54
+ "hidden_chart_oracle_success": 0.52,
55
+ "hidden_chart_oracle_utility": 1.1467878860235214,
56
+ "outcome_ptr": 0.4666666666666667,
57
+ "predicted_margin": 0.33025922453409173,
58
+ "proposal_oracle_success": 0.5333333333333333,
59
+ "proposal_oracle_utility": 1.1093241602430741,
60
+ "selected_candidate_index": 3.8533333333333335,
61
+ "selected_success": 0.30666666666666664,
62
+ "selected_success_gain_over_base": -0.05333333333333334,
63
+ "selected_utility": 0.666174189314867,
64
+ "selected_utility_gain_over_base": -0.10899627190859368,
65
+ "selector_regret": 0.49621246912516653,
66
+ "success_selector_gap": 0.25333333333333335,
67
+ "success_support_gap": 0.02666666666666667,
68
+ "support_gap": 0.1075624164690574,
69
+ "tau": 0.34731881885037885
70
+ },
71
+ "PushCube-v1": {
72
+ "base_success": 0.16666666666666666,
73
+ "base_utility": 0.3924041787783305,
74
+ "coverage": 0.7222222222222222,
75
+ "execute_generated": 0.7222222222222222,
76
+ "fallback_rate": 0.2777777777777778,
77
+ "hidden_chart_oracle_success": 0.8333333333333334,
78
+ "hidden_chart_oracle_utility": 1.7015800600250561,
79
+ "outcome_ptr": 0.8333333333333334,
80
+ "predicted_margin": 0.4544794286452365,
81
+ "proposal_oracle_success": 0.8888888888888888,
82
+ "proposal_oracle_utility": 1.7857814149724112,
83
+ "selected_candidate_index": 4.555555555555555,
84
+ "selected_success": 0.2777777777777778,
85
+ "selected_success_gain_over_base": 0.1111111111111111,
86
+ "selected_utility": 0.6184621560904715,
87
+ "selected_utility_gain_over_base": 0.226057977312141,
88
+ "selector_regret": 1.1673192588819399,
89
+ "success_selector_gap": 0.6111111111111112,
90
+ "success_support_gap": 0.1111111111111111,
91
+ "support_gap": 0.21421858502758873,
92
+ "tau": 0.3473188188503784
93
+ },
94
+ "StackCube-v1": {
95
+ "base_success": 0.2727272727272727,
96
+ "base_utility": 0.7763947045261209,
97
+ "coverage": 0.2727272727272727,
98
+ "execute_generated": 0.2727272727272727,
99
+ "fallback_rate": 0.7272727272727273,
100
+ "hidden_chart_oracle_success": 1.0,
101
+ "hidden_chart_oracle_utility": 2.0,
102
+ "outcome_ptr": 0.6363636363636364,
103
+ "predicted_margin": 0.2862752255853337,
104
+ "proposal_oracle_success": 0.5151515151515151,
105
+ "proposal_oracle_utility": 1.1567951707225856,
106
+ "selected_candidate_index": 6.303030303030303,
107
+ "selected_success": 0.30303030303030304,
108
+ "selected_success_gain_over_base": 0.030303030303030304,
109
+ "selected_utility": 0.7894506924080126,
110
+ "selected_utility_gain_over_base": 0.013055987881891655,
111
+ "selector_regret": 0.37093746346054657,
112
+ "success_selector_gap": 0.21212121212121213,
113
+ "success_support_gap": 0.48484848484848486,
114
+ "support_gap": 0.8432048292774142,
115
+ "tau": 0.3473188188503786
116
+ }
117
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/report.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Nonlinear Train-Calibrated CTT Selector
2
+
3
+ Calibration rows: `432`
4
+ Fit rows: `277`
5
+ Selection rows: `155`
6
+ Eval rows: `144`
7
+ Selected model: `hgb_classifier`
8
+ Selected params: `{"l2_regularization": 0.01, "learning_rate": 0.05, "max_iter": 80, "max_leaf_nodes": 15, "random_state": 0}`
9
+ Tau: `0.347319`
10
+ Feature set: `chart_compat`
11
+ Target: `success_weighted_margin`
12
+
13
+ The model is fit on calibration-fit rows, and model/tau are selected on held-out calibration-selection rows only. Eval outcomes are used only for reporting.
14
+
15
+ | Split | Coverage | Fallback | Base success | Selected success | Proposal oracle | OutcomePTR | Success support gap | Success selector gap |
16
+ | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
17
+ | fit | 0.4549 | 0.5451 | 0.2780 | 0.4621 | 0.5343 | 0.5343 | 0.1733 | 0.1264 |
18
+ | selection | 0.4903 | 0.5097 | 0.2774 | 0.4452 | 0.5419 | 0.5677 | 0.1806 | 0.1419 |
19
+ | eval | 0.3542 | 0.6458 | 0.2917 | 0.2847 | 0.5694 | 0.5486 | 0.2014 | 0.2986 |
20
+
21
+ This is a train-calibrated selector diagnostic over already measured candidates, not a new rollout.
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/split_hash.txt ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "calibration_target_index": {
3
+ "content_hash": "67f7b4a692f7c7e71da24378dc71e3399c5d35afc14c494869eba6087b765c42",
4
+ "retrieval_index_allowed": true,
5
+ "split": "train",
6
+ "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372"
7
+ },
8
+ "eval_target_index": {
9
+ "content_hash": "0c38d73dc9bc4ab756e12006671b5d12f1a23358b281e54b3f6a9ae3cbf6e9af",
10
+ "retrieval_index_allowed": false,
11
+ "split": "test",
12
+ "split_hash": "84151df9e88c219b76751677558cc5bbbc19082bf358cea7330158e9e403e25f"
13
+ },
14
+ "source_index": {
15
+ "content_hash": "67f7b4a692f7c7e71da24378dc71e3399c5d35afc14c494869eba6087b765c42",
16
+ "retrieval_index_allowed": true,
17
+ "split": "train",
18
+ "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372"
19
+ }
20
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/table.tex ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ % Auto-generated by scripts/eval_nonlinear_dominance_selector.py
2
+ \begin{tabular}{lrrrrrrrr}
3
+ \toprule
4
+ Rows & Coverage & Fallback & Base succ. & Selected succ. & Oracle succ. & OutcomePTR & Succ. support gap & Succ. selector gap \\
5
+ \midrule
6
+ 144 & 0.3542 & 0.6458 & 0.2917 & 0.2847 & 0.5694 & 0.5486 & 0.2014 & 0.2986 \\
7
+ \bottomrule
8
+ \end{tabular}
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_envclip_k16_train_to_test/train.log ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ trained nonlinear dominance selector on calibration-fit rows only
2
+ selected_model_type=hgb_classifier
3
+ selected_model_params={"l2_regularization": 0.01, "learning_rate": 0.05, "max_iter": 80, "max_leaf_nodes": 15, "random_state": 0}
4
+ tau=0.347319
5
+ calibration_input=runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/command.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ python scripts/eval_nonlinear_dominance_selector.py --calibration-input runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json --calibration-target-index data/cil_charts_rgb_refs/train/index.json --eval-input runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json --eval-target-index data/cil_charts_rgb_refs/test/index.json --source-index data/cil_charts_rgb_refs/train/index.json --checkpoint-template runs/ctt_residual_base_context_obs_seed{seed}/model.pt --out-dir runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test --feature-set chart_compat --selector-chart-feature-mode base_context_obs --target success_weighted_margin --model-types hgb_regressor,rf_regressor --k 16 --bootstrap-samples 1000
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/config.yaml ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ bootstrap_samples: 1000
2
+ calibration_input: runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json
3
+ calibration_target_index: data/cil_charts_rgb_refs/train/index.json
4
+ checkpoint_template: runs/ctt_residual_base_context_obs_seed{seed}/model.pt
5
+ eval_input: runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json
6
+ eval_target_index: data/cil_charts_rgb_refs/test/index.json
7
+ feature_set: chart_compat
8
+ k: 16
9
+ model_types: hgb_regressor,rf_regressor
10
+ out_dir: runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test
11
+ seed: 0
12
+ selection_frac: 0.35
13
+ selector_chart_feature_mode: base_context_obs
14
+ source_index: data/cil_charts_rgb_refs/train/index.json
15
+ target: success_weighted_margin
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/data_hash.txt ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "calibration_input": "d79e275b896eb98caa128470d4614b76960cab12407faa157c583224be7abe80",
3
+ "calibration_target_index": "67973c50072aa87a3b97533b6292b08a8b4d88a95571dc22e91b6e8fd19e0d85",
4
+ "eval_input": "541d5a928da5f7ee04ca3dea570e04208830dffde431edfe5b5e38b692634045",
5
+ "eval_target_index": "dee1b5225485df6624747dd4c5074bcf514685b54017d05e698d2051638fbf36",
6
+ "source_index": "67973c50072aa87a3b97533b6292b08a8b4d88a95571dc22e91b6e8fd19e0d85"
7
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/eval.log ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ selected model/tau on held-out calibration-selection rows; evaluated on held-out measured rollout rows
2
+ eval_input=runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json
3
+ num_eval_rows=144
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/git_hash.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ 14e81292e2004925f81e0bb4810f6f144bfd87b7
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/metrics.json ADDED
The diff for this file is too large to render. See raw diff
 
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/metrics_by_seed.json ADDED
@@ -0,0 +1,2762 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "1073161778/train0": {
3
+ "base_success": 1.0,
4
+ "base_utility": 2.0,
5
+ "coverage": 0.0,
6
+ "execute_generated": 0.0,
7
+ "fallback_rate": 1.0,
8
+ "hidden_chart_oracle_success": 1.0,
9
+ "hidden_chart_oracle_utility": 2.0,
10
+ "outcome_ptr": 0.0,
11
+ "predicted_margin": 0.06754539929905881,
12
+ "proposal_oracle_success": 1.0,
13
+ "proposal_oracle_utility": 2.0,
14
+ "selected_candidate_index": 9.0,
15
+ "selected_success": 1.0,
16
+ "selected_success_gain_over_base": 0.0,
17
+ "selected_utility": 2.0,
18
+ "selected_utility_gain_over_base": 0.0,
19
+ "selector_regret": 0.0,
20
+ "success_selector_gap": 0.0,
21
+ "success_support_gap": 0.0,
22
+ "support_gap": 0.0,
23
+ "tau": 0.461885278016425
24
+ },
25
+ "1073161778/train1": {
26
+ "base_success": 1.0,
27
+ "base_utility": 2.0,
28
+ "coverage": 0.0,
29
+ "execute_generated": 0.0,
30
+ "fallback_rate": 1.0,
31
+ "hidden_chart_oracle_success": 1.0,
32
+ "hidden_chart_oracle_utility": 2.0,
33
+ "outcome_ptr": 0.0,
34
+ "predicted_margin": 0.06754539929905881,
35
+ "proposal_oracle_success": 1.0,
36
+ "proposal_oracle_utility": 2.0,
37
+ "selected_candidate_index": 1.0,
38
+ "selected_success": 1.0,
39
+ "selected_success_gain_over_base": 0.0,
40
+ "selected_utility": 2.0,
41
+ "selected_utility_gain_over_base": 0.0,
42
+ "selector_regret": 0.0,
43
+ "success_selector_gap": 0.0,
44
+ "success_support_gap": 0.0,
45
+ "support_gap": 0.0,
46
+ "tau": 0.461885278016425
47
+ },
48
+ "1073161778/train2": {
49
+ "base_success": 1.0,
50
+ "base_utility": 2.0,
51
+ "coverage": 0.0,
52
+ "execute_generated": 0.0,
53
+ "fallback_rate": 1.0,
54
+ "hidden_chart_oracle_success": 1.0,
55
+ "hidden_chart_oracle_utility": 2.0,
56
+ "outcome_ptr": 0.0,
57
+ "predicted_margin": 0.06754539929905881,
58
+ "proposal_oracle_success": 1.0,
59
+ "proposal_oracle_utility": 2.0,
60
+ "selected_candidate_index": 4.0,
61
+ "selected_success": 1.0,
62
+ "selected_success_gain_over_base": 0.0,
63
+ "selected_utility": 2.0,
64
+ "selected_utility_gain_over_base": 0.0,
65
+ "selector_regret": 0.0,
66
+ "success_selector_gap": 0.0,
67
+ "success_support_gap": 0.0,
68
+ "support_gap": 0.0,
69
+ "tau": 0.461885278016425
70
+ },
71
+ "1113371261/train0": {
72
+ "base_success": 0.0,
73
+ "base_utility": 0.15842342376708984,
74
+ "coverage": 0.0,
75
+ "execute_generated": 0.0,
76
+ "fallback_rate": 1.0,
77
+ "hidden_chart_oracle_success": 1.0,
78
+ "hidden_chart_oracle_utility": 2.0,
79
+ "outcome_ptr": 0.0,
80
+ "predicted_margin": 0.01069560631150106,
81
+ "proposal_oracle_success": 0.0,
82
+ "proposal_oracle_utility": 0.10274861752986908,
83
+ "selected_candidate_index": 0.0,
84
+ "selected_success": 0.0,
85
+ "selected_success_gain_over_base": 0.0,
86
+ "selected_utility": 0.15842342376708984,
87
+ "selected_utility_gain_over_base": 0.0,
88
+ "selector_regret": 0.0,
89
+ "success_selector_gap": 0.0,
90
+ "success_support_gap": 1.0,
91
+ "support_gap": 1.897251382470131,
92
+ "tau": 0.461885278016425
93
+ },
94
+ "1113371261/train1": {
95
+ "base_success": 0.0,
96
+ "base_utility": 0.15842342376708984,
97
+ "coverage": 0.0,
98
+ "execute_generated": 0.0,
99
+ "fallback_rate": 1.0,
100
+ "hidden_chart_oracle_success": 1.0,
101
+ "hidden_chart_oracle_utility": 2.0,
102
+ "outcome_ptr": 1.0,
103
+ "predicted_margin": 0.03128933348441161,
104
+ "proposal_oracle_success": 0.0,
105
+ "proposal_oracle_utility": 0.18093222379684448,
106
+ "selected_candidate_index": 5.0,
107
+ "selected_success": 0.0,
108
+ "selected_success_gain_over_base": 0.0,
109
+ "selected_utility": 0.15842342376708984,
110
+ "selected_utility_gain_over_base": 0.0,
111
+ "selector_regret": 0.02250880002975464,
112
+ "success_selector_gap": 0.0,
113
+ "success_support_gap": 1.0,
114
+ "support_gap": 1.8190677762031555,
115
+ "tau": 0.461885278016425
116
+ },
117
+ "1113371261/train2": {
118
+ "base_success": 0.0,
119
+ "base_utility": 0.15842342376708984,
120
+ "coverage": 0.0,
121
+ "execute_generated": 0.0,
122
+ "fallback_rate": 1.0,
123
+ "hidden_chart_oracle_success": 1.0,
124
+ "hidden_chart_oracle_utility": 2.0,
125
+ "outcome_ptr": 1.0,
126
+ "predicted_margin": 0.01069560631150106,
127
+ "proposal_oracle_success": 0.0,
128
+ "proposal_oracle_utility": 0.19527167081832886,
129
+ "selected_candidate_index": 2.0,
130
+ "selected_success": 0.0,
131
+ "selected_success_gain_over_base": 0.0,
132
+ "selected_utility": 0.15842342376708984,
133
+ "selected_utility_gain_over_base": 0.0,
134
+ "selector_regret": 0.036848247051239014,
135
+ "success_selector_gap": 0.0,
136
+ "success_support_gap": 1.0,
137
+ "support_gap": 1.8047283291816711,
138
+ "tau": 0.461885278016425
139
+ },
140
+ "1151410346/train0": {
141
+ "base_success": 0.0,
142
+ "base_utility": 0.2260012924671173,
143
+ "coverage": 0.0,
144
+ "execute_generated": 0.0,
145
+ "fallback_rate": 1.0,
146
+ "hidden_chart_oracle_success": 1.0,
147
+ "hidden_chart_oracle_utility": 2.0,
148
+ "outcome_ptr": 1.0,
149
+ "predicted_margin": 0.042155400349067146,
150
+ "proposal_oracle_success": 0.0,
151
+ "proposal_oracle_utility": 0.8663521409034729,
152
+ "selected_candidate_index": 7.0,
153
+ "selected_success": 0.0,
154
+ "selected_success_gain_over_base": 0.0,
155
+ "selected_utility": 0.2260012924671173,
156
+ "selected_utility_gain_over_base": 0.0,
157
+ "selector_regret": 0.6403508484363556,
158
+ "success_selector_gap": 0.0,
159
+ "success_support_gap": 1.0,
160
+ "support_gap": 1.133647859096527,
161
+ "tau": 0.461885278016425
162
+ },
163
+ "1151410346/train1": {
164
+ "base_success": 0.0,
165
+ "base_utility": 0.2260012924671173,
166
+ "coverage": 0.0,
167
+ "execute_generated": 0.0,
168
+ "fallback_rate": 1.0,
169
+ "hidden_chart_oracle_success": 1.0,
170
+ "hidden_chart_oracle_utility": 2.0,
171
+ "outcome_ptr": 1.0,
172
+ "predicted_margin": 0.042155400349067146,
173
+ "proposal_oracle_success": 1.0,
174
+ "proposal_oracle_utility": 2.0,
175
+ "selected_candidate_index": 9.0,
176
+ "selected_success": 0.0,
177
+ "selected_success_gain_over_base": 0.0,
178
+ "selected_utility": 0.2260012924671173,
179
+ "selected_utility_gain_over_base": 0.0,
180
+ "selector_regret": 1.7739987075328827,
181
+ "success_selector_gap": 1.0,
182
+ "success_support_gap": 0.0,
183
+ "support_gap": 0.0,
184
+ "tau": 0.461885278016425
185
+ },
186
+ "1151410346/train2": {
187
+ "base_success": 0.0,
188
+ "base_utility": 0.2260012924671173,
189
+ "coverage": 0.0,
190
+ "execute_generated": 0.0,
191
+ "fallback_rate": 1.0,
192
+ "hidden_chart_oracle_success": 1.0,
193
+ "hidden_chart_oracle_utility": 2.0,
194
+ "outcome_ptr": 1.0,
195
+ "predicted_margin": 0.042155400349067146,
196
+ "proposal_oracle_success": 0.0,
197
+ "proposal_oracle_utility": 0.5522047281265259,
198
+ "selected_candidate_index": 2.0,
199
+ "selected_success": 0.0,
200
+ "selected_success_gain_over_base": 0.0,
201
+ "selected_utility": 0.2260012924671173,
202
+ "selected_utility_gain_over_base": 0.0,
203
+ "selector_regret": 0.32620343565940857,
204
+ "success_selector_gap": 0.0,
205
+ "success_support_gap": 1.0,
206
+ "support_gap": 1.4477952718734741,
207
+ "tau": 0.461885278016425
208
+ },
209
+ "1261930456/train0": {
210
+ "base_success": 1.0,
211
+ "base_utility": 2.0,
212
+ "coverage": 0.0,
213
+ "execute_generated": 0.0,
214
+ "fallback_rate": 1.0,
215
+ "hidden_chart_oracle_success": 1.0,
216
+ "hidden_chart_oracle_utility": 2.0,
217
+ "outcome_ptr": 0.0,
218
+ "predicted_margin": 0.09105107967198207,
219
+ "proposal_oracle_success": 1.0,
220
+ "proposal_oracle_utility": 2.0,
221
+ "selected_candidate_index": 0.0,
222
+ "selected_success": 1.0,
223
+ "selected_success_gain_over_base": 0.0,
224
+ "selected_utility": 2.0,
225
+ "selected_utility_gain_over_base": 0.0,
226
+ "selector_regret": 0.0,
227
+ "success_selector_gap": 0.0,
228
+ "success_support_gap": 0.0,
229
+ "support_gap": 0.0,
230
+ "tau": 0.461885278016425
231
+ },
232
+ "1261930456/train1": {
233
+ "base_success": 1.0,
234
+ "base_utility": 2.0,
235
+ "coverage": 0.0,
236
+ "execute_generated": 0.0,
237
+ "fallback_rate": 1.0,
238
+ "hidden_chart_oracle_success": 1.0,
239
+ "hidden_chart_oracle_utility": 2.0,
240
+ "outcome_ptr": 0.0,
241
+ "predicted_margin": 0.1179583821973945,
242
+ "proposal_oracle_success": 1.0,
243
+ "proposal_oracle_utility": 2.0,
244
+ "selected_candidate_index": 10.0,
245
+ "selected_success": 1.0,
246
+ "selected_success_gain_over_base": 0.0,
247
+ "selected_utility": 2.0,
248
+ "selected_utility_gain_over_base": 0.0,
249
+ "selector_regret": 0.0,
250
+ "success_selector_gap": 0.0,
251
+ "success_support_gap": 0.0,
252
+ "support_gap": 0.0,
253
+ "tau": 0.461885278016425
254
+ },
255
+ "1261930456/train2": {
256
+ "base_success": 1.0,
257
+ "base_utility": 2.0,
258
+ "coverage": 0.0,
259
+ "execute_generated": 0.0,
260
+ "fallback_rate": 1.0,
261
+ "hidden_chart_oracle_success": 1.0,
262
+ "hidden_chart_oracle_utility": 2.0,
263
+ "outcome_ptr": 0.0,
264
+ "predicted_margin": 0.09592156335756839,
265
+ "proposal_oracle_success": 1.0,
266
+ "proposal_oracle_utility": 2.0,
267
+ "selected_candidate_index": 4.0,
268
+ "selected_success": 1.0,
269
+ "selected_success_gain_over_base": 0.0,
270
+ "selected_utility": 2.0,
271
+ "selected_utility_gain_over_base": 0.0,
272
+ "selector_regret": 0.0,
273
+ "success_selector_gap": 0.0,
274
+ "success_support_gap": 0.0,
275
+ "support_gap": 0.0,
276
+ "tau": 0.461885278016425
277
+ },
278
+ "1324133122/train0": {
279
+ "base_success": 0.0,
280
+ "base_utility": 0.09220491349697113,
281
+ "coverage": 0.0,
282
+ "execute_generated": 0.0,
283
+ "fallback_rate": 1.0,
284
+ "hidden_chart_oracle_success": 1.0,
285
+ "hidden_chart_oracle_utility": 2.0,
286
+ "outcome_ptr": 0.0,
287
+ "predicted_margin": 0.07276587335777485,
288
+ "proposal_oracle_success": 0.0,
289
+ "proposal_oracle_utility": 0.054989203810691833,
290
+ "selected_candidate_index": 0.0,
291
+ "selected_success": 0.0,
292
+ "selected_success_gain_over_base": 0.0,
293
+ "selected_utility": 0.09220491349697113,
294
+ "selected_utility_gain_over_base": 0.0,
295
+ "selector_regret": 0.0,
296
+ "success_selector_gap": 0.0,
297
+ "success_support_gap": 1.0,
298
+ "support_gap": 1.9450107961893082,
299
+ "tau": 0.461885278016425
300
+ },
301
+ "1324133122/train1": {
302
+ "base_success": 0.0,
303
+ "base_utility": 0.09220491349697113,
304
+ "coverage": 0.0,
305
+ "execute_generated": 0.0,
306
+ "fallback_rate": 1.0,
307
+ "hidden_chart_oracle_success": 1.0,
308
+ "hidden_chart_oracle_utility": 2.0,
309
+ "outcome_ptr": 1.0,
310
+ "predicted_margin": 0.07276587335777485,
311
+ "proposal_oracle_success": 0.0,
312
+ "proposal_oracle_utility": 0.1324586570262909,
313
+ "selected_candidate_index": 2.0,
314
+ "selected_success": 0.0,
315
+ "selected_success_gain_over_base": 0.0,
316
+ "selected_utility": 0.09220491349697113,
317
+ "selected_utility_gain_over_base": 0.0,
318
+ "selector_regret": 0.04025374352931976,
319
+ "success_selector_gap": 0.0,
320
+ "success_support_gap": 1.0,
321
+ "support_gap": 1.867541342973709,
322
+ "tau": 0.461885278016425
323
+ },
324
+ "1324133122/train2": {
325
+ "base_success": 0.0,
326
+ "base_utility": 0.09220491349697113,
327
+ "coverage": 0.0,
328
+ "execute_generated": 0.0,
329
+ "fallback_rate": 1.0,
330
+ "hidden_chart_oracle_success": 1.0,
331
+ "hidden_chart_oracle_utility": 2.0,
332
+ "outcome_ptr": 1.0,
333
+ "predicted_margin": 0.07276587335777485,
334
+ "proposal_oracle_success": 0.0,
335
+ "proposal_oracle_utility": 0.19291165471076965,
336
+ "selected_candidate_index": 1.0,
337
+ "selected_success": 0.0,
338
+ "selected_success_gain_over_base": 0.0,
339
+ "selected_utility": 0.09220491349697113,
340
+ "selected_utility_gain_over_base": 0.0,
341
+ "selector_regret": 0.10070674121379852,
342
+ "success_selector_gap": 0.0,
343
+ "success_support_gap": 1.0,
344
+ "support_gap": 1.8070883452892303,
345
+ "tau": 0.461885278016425
346
+ },
347
+ "1324642411/train0": {
348
+ "base_success": 0.0,
349
+ "base_utility": 0.10647052526473999,
350
+ "coverage": 1.0,
351
+ "execute_generated": 1.0,
352
+ "fallback_rate": 0.0,
353
+ "hidden_chart_oracle_success": 1.0,
354
+ "hidden_chart_oracle_utility": 2.0,
355
+ "outcome_ptr": 1.0,
356
+ "predicted_margin": 0.5083763279916524,
357
+ "proposal_oracle_success": 0.0,
358
+ "proposal_oracle_utility": 0.14406004548072815,
359
+ "selected_candidate_index": 12.0,
360
+ "selected_success": 0.0,
361
+ "selected_success_gain_over_base": 0.0,
362
+ "selected_utility": 0.14406004548072815,
363
+ "selected_utility_gain_over_base": 0.03758952021598816,
364
+ "selector_regret": 0.0,
365
+ "success_selector_gap": 0.0,
366
+ "success_support_gap": 1.0,
367
+ "support_gap": 1.8559399545192719,
368
+ "tau": 0.461885278016425
369
+ },
370
+ "1324642411/train1": {
371
+ "base_success": 0.0,
372
+ "base_utility": 0.10647052526473999,
373
+ "coverage": 1.0,
374
+ "execute_generated": 1.0,
375
+ "fallback_rate": 0.0,
376
+ "hidden_chart_oracle_success": 1.0,
377
+ "hidden_chart_oracle_utility": 2.0,
378
+ "outcome_ptr": 1.0,
379
+ "predicted_margin": 0.5702108833985109,
380
+ "proposal_oracle_success": 1.0,
381
+ "proposal_oracle_utility": 2.0,
382
+ "selected_candidate_index": 2.0,
383
+ "selected_success": 0.0,
384
+ "selected_success_gain_over_base": 0.0,
385
+ "selected_utility": 0.13300009071826935,
386
+ "selected_utility_gain_over_base": 0.026529565453529358,
387
+ "selector_regret": 1.8669999092817307,
388
+ "success_selector_gap": 1.0,
389
+ "success_support_gap": 0.0,
390
+ "support_gap": 0.0,
391
+ "tau": 0.461885278016425
392
+ },
393
+ "1324642411/train2": {
394
+ "base_success": 0.0,
395
+ "base_utility": 0.10647052526473999,
396
+ "coverage": 1.0,
397
+ "execute_generated": 1.0,
398
+ "fallback_rate": 0.0,
399
+ "hidden_chart_oracle_success": 1.0,
400
+ "hidden_chart_oracle_utility": 2.0,
401
+ "outcome_ptr": 1.0,
402
+ "predicted_margin": 0.49404811032777285,
403
+ "proposal_oracle_success": 1.0,
404
+ "proposal_oracle_utility": 2.0,
405
+ "selected_candidate_index": 0.0,
406
+ "selected_success": 0.0,
407
+ "selected_success_gain_over_base": 0.0,
408
+ "selected_utility": 0.11692000925540924,
409
+ "selected_utility_gain_over_base": 0.01044948399066925,
410
+ "selector_regret": 1.8830799907445908,
411
+ "success_selector_gap": 1.0,
412
+ "success_support_gap": 0.0,
413
+ "support_gap": 0.0,
414
+ "tau": 0.461885278016425
415
+ },
416
+ "1375492786/train0": {
417
+ "base_success": 1.0,
418
+ "base_utility": 2.0,
419
+ "coverage": 0.0,
420
+ "execute_generated": 0.0,
421
+ "fallback_rate": 1.0,
422
+ "hidden_chart_oracle_success": 1.0,
423
+ "hidden_chart_oracle_utility": 2.0,
424
+ "outcome_ptr": 0.0,
425
+ "predicted_margin": -0.03964445208865493,
426
+ "proposal_oracle_success": 1.0,
427
+ "proposal_oracle_utility": 2.0,
428
+ "selected_candidate_index": 10.0,
429
+ "selected_success": 1.0,
430
+ "selected_success_gain_over_base": 0.0,
431
+ "selected_utility": 2.0,
432
+ "selected_utility_gain_over_base": 0.0,
433
+ "selector_regret": 0.0,
434
+ "success_selector_gap": 0.0,
435
+ "success_support_gap": 0.0,
436
+ "support_gap": 0.0,
437
+ "tau": 0.461885278016425
438
+ },
439
+ "1375492786/train1": {
440
+ "base_success": 1.0,
441
+ "base_utility": 2.0,
442
+ "coverage": 0.0,
443
+ "execute_generated": 0.0,
444
+ "fallback_rate": 1.0,
445
+ "hidden_chart_oracle_success": 1.0,
446
+ "hidden_chart_oracle_utility": 2.0,
447
+ "outcome_ptr": 0.0,
448
+ "predicted_margin": 0.019438254631384468,
449
+ "proposal_oracle_success": 1.0,
450
+ "proposal_oracle_utility": 2.0,
451
+ "selected_candidate_index": 1.0,
452
+ "selected_success": 1.0,
453
+ "selected_success_gain_over_base": 0.0,
454
+ "selected_utility": 2.0,
455
+ "selected_utility_gain_over_base": 0.0,
456
+ "selector_regret": 0.0,
457
+ "success_selector_gap": 0.0,
458
+ "success_support_gap": 0.0,
459
+ "support_gap": 0.0,
460
+ "tau": 0.461885278016425
461
+ },
462
+ "1375492786/train2": {
463
+ "base_success": 1.0,
464
+ "base_utility": 2.0,
465
+ "coverage": 0.0,
466
+ "execute_generated": 0.0,
467
+ "fallback_rate": 1.0,
468
+ "hidden_chart_oracle_success": 1.0,
469
+ "hidden_chart_oracle_utility": 2.0,
470
+ "outcome_ptr": 0.0,
471
+ "predicted_margin": 0.019438254631384468,
472
+ "proposal_oracle_success": 1.0,
473
+ "proposal_oracle_utility": 2.0,
474
+ "selected_candidate_index": 2.0,
475
+ "selected_success": 1.0,
476
+ "selected_success_gain_over_base": 0.0,
477
+ "selected_utility": 2.0,
478
+ "selected_utility_gain_over_base": 0.0,
479
+ "selector_regret": 0.0,
480
+ "success_selector_gap": 0.0,
481
+ "success_support_gap": 0.0,
482
+ "support_gap": 0.0,
483
+ "tau": 0.461885278016425
484
+ },
485
+ "141984615/train0": {
486
+ "base_success": 0.0,
487
+ "base_utility": 0.10283424705266953,
488
+ "coverage": 0.0,
489
+ "execute_generated": 0.0,
490
+ "fallback_rate": 1.0,
491
+ "hidden_chart_oracle_success": 0.0,
492
+ "hidden_chart_oracle_utility": 0.2857772707939148,
493
+ "outcome_ptr": 0.0,
494
+ "predicted_margin": 0.45076360899263673,
495
+ "proposal_oracle_success": 0.0,
496
+ "proposal_oracle_utility": 0.015177905559539795,
497
+ "selected_candidate_index": 10.0,
498
+ "selected_success": 0.0,
499
+ "selected_success_gain_over_base": 0.0,
500
+ "selected_utility": 0.10283424705266953,
501
+ "selected_utility_gain_over_base": 0.0,
502
+ "selector_regret": 0.0,
503
+ "success_selector_gap": 0.0,
504
+ "success_support_gap": 0.0,
505
+ "support_gap": 0.270599365234375,
506
+ "tau": 0.461885278016425
507
+ },
508
+ "141984615/train1": {
509
+ "base_success": 0.0,
510
+ "base_utility": 0.10283424705266953,
511
+ "coverage": 1.0,
512
+ "execute_generated": 1.0,
513
+ "fallback_rate": 0.0,
514
+ "hidden_chart_oracle_success": 0.0,
515
+ "hidden_chart_oracle_utility": 0.2857772707939148,
516
+ "outcome_ptr": 0.0,
517
+ "predicted_margin": 0.6018704158695247,
518
+ "proposal_oracle_success": 0.0,
519
+ "proposal_oracle_utility": 0.037011127918958664,
520
+ "selected_candidate_index": 1.0,
521
+ "selected_success": 0.0,
522
+ "selected_success_gain_over_base": 0.0,
523
+ "selected_utility": 0.01966226100921631,
524
+ "selected_utility_gain_over_base": -0.08317198604345322,
525
+ "selector_regret": 0.017348866909742355,
526
+ "success_selector_gap": 0.0,
527
+ "success_support_gap": 0.0,
528
+ "support_gap": 0.24876614287495613,
529
+ "tau": 0.461885278016425
530
+ },
531
+ "141984615/train2": {
532
+ "base_success": 0.0,
533
+ "base_utility": 0.10283424705266953,
534
+ "coverage": 0.0,
535
+ "execute_generated": 0.0,
536
+ "fallback_rate": 1.0,
537
+ "hidden_chart_oracle_success": 0.0,
538
+ "hidden_chart_oracle_utility": 0.2857772707939148,
539
+ "outcome_ptr": 0.0,
540
+ "predicted_margin": 0.45076360899263673,
541
+ "proposal_oracle_success": 0.0,
542
+ "proposal_oracle_utility": 0.058951716870069504,
543
+ "selected_candidate_index": 11.0,
544
+ "selected_success": 0.0,
545
+ "selected_success_gain_over_base": 0.0,
546
+ "selected_utility": 0.10283424705266953,
547
+ "selected_utility_gain_over_base": 0.0,
548
+ "selector_regret": 0.0,
549
+ "success_selector_gap": 0.0,
550
+ "success_support_gap": 0.0,
551
+ "support_gap": 0.2268255539238453,
552
+ "tau": 0.461885278016425
553
+ },
554
+ "1461069713/train0": {
555
+ "base_success": 0.0,
556
+ "base_utility": 0.0939030647277832,
557
+ "coverage": 0.0,
558
+ "execute_generated": 0.0,
559
+ "fallback_rate": 1.0,
560
+ "hidden_chart_oracle_success": 0.0,
561
+ "hidden_chart_oracle_utility": 0.1593424528837204,
562
+ "outcome_ptr": 0.0,
563
+ "predicted_margin": 0.030876378763120833,
564
+ "proposal_oracle_success": 0.0,
565
+ "proposal_oracle_utility": 0.06019836664199829,
566
+ "selected_candidate_index": 2.0,
567
+ "selected_success": 0.0,
568
+ "selected_success_gain_over_base": 0.0,
569
+ "selected_utility": 0.0939030647277832,
570
+ "selected_utility_gain_over_base": 0.0,
571
+ "selector_regret": 0.0,
572
+ "success_selector_gap": 0.0,
573
+ "success_support_gap": 0.0,
574
+ "support_gap": 0.0991440862417221,
575
+ "tau": 0.461885278016425
576
+ },
577
+ "1461069713/train1": {
578
+ "base_success": 0.0,
579
+ "base_utility": 0.0939030647277832,
580
+ "coverage": 0.0,
581
+ "execute_generated": 0.0,
582
+ "fallback_rate": 1.0,
583
+ "hidden_chart_oracle_success": 0.0,
584
+ "hidden_chart_oracle_utility": 0.1593424528837204,
585
+ "outcome_ptr": 0.0,
586
+ "predicted_margin": 0.05928441239142939,
587
+ "proposal_oracle_success": 0.0,
588
+ "proposal_oracle_utility": 0.016585012897849083,
589
+ "selected_candidate_index": 0.0,
590
+ "selected_success": 0.0,
591
+ "selected_success_gain_over_base": 0.0,
592
+ "selected_utility": 0.0939030647277832,
593
+ "selected_utility_gain_over_base": 0.0,
594
+ "selector_regret": 0.0,
595
+ "success_selector_gap": 0.0,
596
+ "success_support_gap": 0.0,
597
+ "support_gap": 0.14275743998587132,
598
+ "tau": 0.461885278016425
599
+ },
600
+ "1461069713/train2": {
601
+ "base_success": 0.0,
602
+ "base_utility": 0.0939030647277832,
603
+ "coverage": 0.0,
604
+ "execute_generated": 0.0,
605
+ "fallback_rate": 1.0,
606
+ "hidden_chart_oracle_success": 0.0,
607
+ "hidden_chart_oracle_utility": 0.1593424528837204,
608
+ "outcome_ptr": 1.0,
609
+ "predicted_margin": 0.030876378763120833,
610
+ "proposal_oracle_success": 0.0,
611
+ "proposal_oracle_utility": 0.09997648000717163,
612
+ "selected_candidate_index": 5.0,
613
+ "selected_success": 0.0,
614
+ "selected_success_gain_over_base": 0.0,
615
+ "selected_utility": 0.0939030647277832,
616
+ "selected_utility_gain_over_base": 0.0,
617
+ "selector_regret": 0.006073415279388428,
618
+ "success_selector_gap": 0.0,
619
+ "success_support_gap": 0.0,
620
+ "support_gap": 0.05936597287654877,
621
+ "tau": 0.461885278016425
622
+ },
623
+ "1480856810/train0": {
624
+ "base_success": 0.0,
625
+ "base_utility": 0.08703407645225525,
626
+ "coverage": 0.0,
627
+ "execute_generated": 0.0,
628
+ "fallback_rate": 1.0,
629
+ "hidden_chart_oracle_success": 0.0,
630
+ "hidden_chart_oracle_utility": 0.20948036015033722,
631
+ "outcome_ptr": 1.0,
632
+ "predicted_margin": 0.23502340911767072,
633
+ "proposal_oracle_success": 1.0,
634
+ "proposal_oracle_utility": 2.0,
635
+ "selected_candidate_index": 2.0,
636
+ "selected_success": 0.0,
637
+ "selected_success_gain_over_base": 0.0,
638
+ "selected_utility": 0.08703407645225525,
639
+ "selected_utility_gain_over_base": 0.0,
640
+ "selector_regret": 1.9129659235477448,
641
+ "success_selector_gap": 1.0,
642
+ "success_support_gap": 0.0,
643
+ "support_gap": 0.0,
644
+ "tau": 0.461885278016425
645
+ },
646
+ "1480856810/train1": {
647
+ "base_success": 0.0,
648
+ "base_utility": 0.08703407645225525,
649
+ "coverage": 0.0,
650
+ "execute_generated": 0.0,
651
+ "fallback_rate": 1.0,
652
+ "hidden_chart_oracle_success": 0.0,
653
+ "hidden_chart_oracle_utility": 0.20948036015033722,
654
+ "outcome_ptr": 1.0,
655
+ "predicted_margin": 0.32219394509834176,
656
+ "proposal_oracle_success": 1.0,
657
+ "proposal_oracle_utility": 2.0,
658
+ "selected_candidate_index": 13.0,
659
+ "selected_success": 0.0,
660
+ "selected_success_gain_over_base": 0.0,
661
+ "selected_utility": 0.08703407645225525,
662
+ "selected_utility_gain_over_base": 0.0,
663
+ "selector_regret": 1.9129659235477448,
664
+ "success_selector_gap": 1.0,
665
+ "success_support_gap": 0.0,
666
+ "support_gap": 0.0,
667
+ "tau": 0.461885278016425
668
+ },
669
+ "1480856810/train2": {
670
+ "base_success": 0.0,
671
+ "base_utility": 0.08703407645225525,
672
+ "coverage": 0.0,
673
+ "execute_generated": 0.0,
674
+ "fallback_rate": 1.0,
675
+ "hidden_chart_oracle_success": 0.0,
676
+ "hidden_chart_oracle_utility": 0.20948036015033722,
677
+ "outcome_ptr": 1.0,
678
+ "predicted_margin": 0.26627686358563757,
679
+ "proposal_oracle_success": 1.0,
680
+ "proposal_oracle_utility": 2.0,
681
+ "selected_candidate_index": 1.0,
682
+ "selected_success": 0.0,
683
+ "selected_success_gain_over_base": 0.0,
684
+ "selected_utility": 0.08703407645225525,
685
+ "selected_utility_gain_over_base": 0.0,
686
+ "selector_regret": 1.9129659235477448,
687
+ "success_selector_gap": 1.0,
688
+ "success_support_gap": 0.0,
689
+ "support_gap": 0.0,
690
+ "tau": 0.461885278016425
691
+ },
692
+ "1509980664/train0": {
693
+ "base_success": 0.0,
694
+ "base_utility": 0.8334100246429443,
695
+ "coverage": 1.0,
696
+ "execute_generated": 1.0,
697
+ "fallback_rate": 0.0,
698
+ "hidden_chart_oracle_success": 1.0,
699
+ "hidden_chart_oracle_utility": 2.0,
700
+ "outcome_ptr": 1.0,
701
+ "predicted_margin": 0.5976092755124571,
702
+ "proposal_oracle_success": 1.0,
703
+ "proposal_oracle_utility": 2.0,
704
+ "selected_candidate_index": 10.0,
705
+ "selected_success": 1.0,
706
+ "selected_success_gain_over_base": 1.0,
707
+ "selected_utility": 2.0,
708
+ "selected_utility_gain_over_base": 1.1665899753570557,
709
+ "selector_regret": 0.0,
710
+ "success_selector_gap": 0.0,
711
+ "success_support_gap": 0.0,
712
+ "support_gap": 0.0,
713
+ "tau": 0.461885278016425
714
+ },
715
+ "1509980664/train1": {
716
+ "base_success": 0.0,
717
+ "base_utility": 0.8334100246429443,
718
+ "coverage": 1.0,
719
+ "execute_generated": 1.0,
720
+ "fallback_rate": 0.0,
721
+ "hidden_chart_oracle_success": 1.0,
722
+ "hidden_chart_oracle_utility": 2.0,
723
+ "outcome_ptr": 1.0,
724
+ "predicted_margin": 0.5976092755124571,
725
+ "proposal_oracle_success": 1.0,
726
+ "proposal_oracle_utility": 2.0,
727
+ "selected_candidate_index": 9.0,
728
+ "selected_success": 0.0,
729
+ "selected_success_gain_over_base": 0.0,
730
+ "selected_utility": 0.8125686645507812,
731
+ "selected_utility_gain_over_base": -0.020841360092163086,
732
+ "selector_regret": 1.1874313354492188,
733
+ "success_selector_gap": 1.0,
734
+ "success_support_gap": 0.0,
735
+ "support_gap": 0.0,
736
+ "tau": 0.461885278016425
737
+ },
738
+ "1509980664/train2": {
739
+ "base_success": 0.0,
740
+ "base_utility": 0.8334100246429443,
741
+ "coverage": 1.0,
742
+ "execute_generated": 1.0,
743
+ "fallback_rate": 0.0,
744
+ "hidden_chart_oracle_success": 1.0,
745
+ "hidden_chart_oracle_utility": 2.0,
746
+ "outcome_ptr": 1.0,
747
+ "predicted_margin": 0.5976092755124571,
748
+ "proposal_oracle_success": 1.0,
749
+ "proposal_oracle_utility": 2.0,
750
+ "selected_candidate_index": 4.0,
751
+ "selected_success": 0.0,
752
+ "selected_success_gain_over_base": 0.0,
753
+ "selected_utility": 0.18445390462875366,
754
+ "selected_utility_gain_over_base": -0.6489561200141907,
755
+ "selector_regret": 1.8155460953712463,
756
+ "success_selector_gap": 1.0,
757
+ "success_support_gap": 0.0,
758
+ "support_gap": 0.0,
759
+ "tau": 0.461885278016425
760
+ },
761
+ "1522272686/train0": {
762
+ "base_success": 0.0,
763
+ "base_utility": 0.5839591026306152,
764
+ "coverage": 0.0,
765
+ "execute_generated": 0.0,
766
+ "fallback_rate": 1.0,
767
+ "hidden_chart_oracle_success": 1.0,
768
+ "hidden_chart_oracle_utility": 2.0,
769
+ "outcome_ptr": 0.0,
770
+ "predicted_margin": -0.07343996006022091,
771
+ "proposal_oracle_success": 0.0,
772
+ "proposal_oracle_utility": 0.18325337767601013,
773
+ "selected_candidate_index": 7.0,
774
+ "selected_success": 0.0,
775
+ "selected_success_gain_over_base": 0.0,
776
+ "selected_utility": 0.5839591026306152,
777
+ "selected_utility_gain_over_base": 0.0,
778
+ "selector_regret": 0.0,
779
+ "success_selector_gap": 0.0,
780
+ "success_support_gap": 1.0,
781
+ "support_gap": 1.8167466223239899,
782
+ "tau": 0.461885278016425
783
+ },
784
+ "1522272686/train1": {
785
+ "base_success": 0.0,
786
+ "base_utility": 0.5839591026306152,
787
+ "coverage": 0.0,
788
+ "execute_generated": 0.0,
789
+ "fallback_rate": 1.0,
790
+ "hidden_chart_oracle_success": 1.0,
791
+ "hidden_chart_oracle_utility": 2.0,
792
+ "outcome_ptr": 0.0,
793
+ "predicted_margin": -0.07343996006022091,
794
+ "proposal_oracle_success": 0.0,
795
+ "proposal_oracle_utility": 0.07014383375644684,
796
+ "selected_candidate_index": 7.0,
797
+ "selected_success": 0.0,
798
+ "selected_success_gain_over_base": 0.0,
799
+ "selected_utility": 0.5839591026306152,
800
+ "selected_utility_gain_over_base": 0.0,
801
+ "selector_regret": 0.0,
802
+ "success_selector_gap": 0.0,
803
+ "success_support_gap": 1.0,
804
+ "support_gap": 1.9298561662435532,
805
+ "tau": 0.461885278016425
806
+ },
807
+ "1522272686/train2": {
808
+ "base_success": 0.0,
809
+ "base_utility": 0.5839591026306152,
810
+ "coverage": 0.0,
811
+ "execute_generated": 0.0,
812
+ "fallback_rate": 1.0,
813
+ "hidden_chart_oracle_success": 1.0,
814
+ "hidden_chart_oracle_utility": 2.0,
815
+ "outcome_ptr": 0.0,
816
+ "predicted_margin": -0.07343996006022091,
817
+ "proposal_oracle_success": 0.0,
818
+ "proposal_oracle_utility": 0.14937306940555573,
819
+ "selected_candidate_index": 6.0,
820
+ "selected_success": 0.0,
821
+ "selected_success_gain_over_base": 0.0,
822
+ "selected_utility": 0.5839591026306152,
823
+ "selected_utility_gain_over_base": 0.0,
824
+ "selector_regret": 0.0,
825
+ "success_selector_gap": 0.0,
826
+ "success_support_gap": 1.0,
827
+ "support_gap": 1.8506269305944443,
828
+ "tau": 0.461885278016425
829
+ },
830
+ "1549324805/train0": {
831
+ "base_success": 0.0,
832
+ "base_utility": 0.10451483726501465,
833
+ "coverage": 0.0,
834
+ "execute_generated": 0.0,
835
+ "fallback_rate": 1.0,
836
+ "hidden_chart_oracle_success": 0.0,
837
+ "hidden_chart_oracle_utility": 0.2974560856819153,
838
+ "outcome_ptr": 0.0,
839
+ "predicted_margin": 0.04057706719583071,
840
+ "proposal_oracle_success": 0.0,
841
+ "proposal_oracle_utility": 0.060871422290802,
842
+ "selected_candidate_index": 6.0,
843
+ "selected_success": 0.0,
844
+ "selected_success_gain_over_base": 0.0,
845
+ "selected_utility": 0.10451483726501465,
846
+ "selected_utility_gain_over_base": 0.0,
847
+ "selector_regret": 0.0,
848
+ "success_selector_gap": 0.0,
849
+ "success_support_gap": 0.0,
850
+ "support_gap": 0.23658466339111328,
851
+ "tau": 0.461885278016425
852
+ },
853
+ "1549324805/train1": {
854
+ "base_success": 0.0,
855
+ "base_utility": 0.10451483726501465,
856
+ "coverage": 0.0,
857
+ "execute_generated": 0.0,
858
+ "fallback_rate": 1.0,
859
+ "hidden_chart_oracle_success": 0.0,
860
+ "hidden_chart_oracle_utility": 0.2974560856819153,
861
+ "outcome_ptr": 1.0,
862
+ "predicted_margin": 0.04057706719583071,
863
+ "proposal_oracle_success": 0.0,
864
+ "proposal_oracle_utility": 0.19892151653766632,
865
+ "selected_candidate_index": 0.0,
866
+ "selected_success": 0.0,
867
+ "selected_success_gain_over_base": 0.0,
868
+ "selected_utility": 0.10451483726501465,
869
+ "selected_utility_gain_over_base": 0.0,
870
+ "selector_regret": 0.09440667927265167,
871
+ "success_selector_gap": 0.0,
872
+ "success_support_gap": 0.0,
873
+ "support_gap": 0.09853456914424896,
874
+ "tau": 0.461885278016425
875
+ },
876
+ "1549324805/train2": {
877
+ "base_success": 0.0,
878
+ "base_utility": 0.10451483726501465,
879
+ "coverage": 0.0,
880
+ "execute_generated": 0.0,
881
+ "fallback_rate": 1.0,
882
+ "hidden_chart_oracle_success": 0.0,
883
+ "hidden_chart_oracle_utility": 0.2974560856819153,
884
+ "outcome_ptr": 1.0,
885
+ "predicted_margin": 0.04057706719583071,
886
+ "proposal_oracle_success": 0.0,
887
+ "proposal_oracle_utility": 0.11310184001922607,
888
+ "selected_candidate_index": 12.0,
889
+ "selected_success": 0.0,
890
+ "selected_success_gain_over_base": 0.0,
891
+ "selected_utility": 0.10451483726501465,
892
+ "selected_utility_gain_over_base": 0.0,
893
+ "selector_regret": 0.008587002754211426,
894
+ "success_selector_gap": 0.0,
895
+ "success_support_gap": 0.0,
896
+ "support_gap": 0.1843542456626892,
897
+ "tau": 0.461885278016425
898
+ },
899
+ "1578990937/train0": {
900
+ "base_success": 0.0,
901
+ "base_utility": 0.82413649559021,
902
+ "coverage": 0.0,
903
+ "execute_generated": 0.0,
904
+ "fallback_rate": 1.0,
905
+ "hidden_chart_oracle_success": 1.0,
906
+ "hidden_chart_oracle_utility": 2.0,
907
+ "outcome_ptr": 1.0,
908
+ "predicted_margin": 0.14846214977339026,
909
+ "proposal_oracle_success": 1.0,
910
+ "proposal_oracle_utility": 2.0,
911
+ "selected_candidate_index": 0.0,
912
+ "selected_success": 0.0,
913
+ "selected_success_gain_over_base": 0.0,
914
+ "selected_utility": 0.82413649559021,
915
+ "selected_utility_gain_over_base": 0.0,
916
+ "selector_regret": 1.17586350440979,
917
+ "success_selector_gap": 1.0,
918
+ "success_support_gap": 0.0,
919
+ "support_gap": 0.0,
920
+ "tau": 0.461885278016425
921
+ },
922
+ "1578990937/train1": {
923
+ "base_success": 0.0,
924
+ "base_utility": 0.82413649559021,
925
+ "coverage": 0.0,
926
+ "execute_generated": 0.0,
927
+ "fallback_rate": 1.0,
928
+ "hidden_chart_oracle_success": 1.0,
929
+ "hidden_chart_oracle_utility": 2.0,
930
+ "outcome_ptr": 1.0,
931
+ "predicted_margin": 0.14846214977339026,
932
+ "proposal_oracle_success": 1.0,
933
+ "proposal_oracle_utility": 2.0,
934
+ "selected_candidate_index": 1.0,
935
+ "selected_success": 0.0,
936
+ "selected_success_gain_over_base": 0.0,
937
+ "selected_utility": 0.82413649559021,
938
+ "selected_utility_gain_over_base": 0.0,
939
+ "selector_regret": 1.17586350440979,
940
+ "success_selector_gap": 1.0,
941
+ "success_support_gap": 0.0,
942
+ "support_gap": 0.0,
943
+ "tau": 0.461885278016425
944
+ },
945
+ "1578990937/train2": {
946
+ "base_success": 0.0,
947
+ "base_utility": 0.82413649559021,
948
+ "coverage": 0.0,
949
+ "execute_generated": 0.0,
950
+ "fallback_rate": 1.0,
951
+ "hidden_chart_oracle_success": 1.0,
952
+ "hidden_chart_oracle_utility": 2.0,
953
+ "outcome_ptr": 1.0,
954
+ "predicted_margin": 0.14846214977339026,
955
+ "proposal_oracle_success": 1.0,
956
+ "proposal_oracle_utility": 2.0,
957
+ "selected_candidate_index": 1.0,
958
+ "selected_success": 0.0,
959
+ "selected_success_gain_over_base": 0.0,
960
+ "selected_utility": 0.82413649559021,
961
+ "selected_utility_gain_over_base": 0.0,
962
+ "selector_regret": 1.17586350440979,
963
+ "success_selector_gap": 1.0,
964
+ "success_support_gap": 0.0,
965
+ "support_gap": 0.0,
966
+ "tau": 0.461885278016425
967
+ },
968
+ "1747178030/train0": {
969
+ "base_success": 1.0,
970
+ "base_utility": 2.0,
971
+ "coverage": 0.0,
972
+ "execute_generated": 0.0,
973
+ "fallback_rate": 1.0,
974
+ "hidden_chart_oracle_success": 1.0,
975
+ "hidden_chart_oracle_utility": 2.0,
976
+ "outcome_ptr": 0.0,
977
+ "predicted_margin": -0.09430490853290777,
978
+ "proposal_oracle_success": 1.0,
979
+ "proposal_oracle_utility": 2.0,
980
+ "selected_candidate_index": 3.0,
981
+ "selected_success": 1.0,
982
+ "selected_success_gain_over_base": 0.0,
983
+ "selected_utility": 2.0,
984
+ "selected_utility_gain_over_base": 0.0,
985
+ "selector_regret": 0.0,
986
+ "success_selector_gap": 0.0,
987
+ "success_support_gap": 0.0,
988
+ "support_gap": 0.0,
989
+ "tau": 0.461885278016425
990
+ },
991
+ "1747178030/train1": {
992
+ "base_success": 1.0,
993
+ "base_utility": 2.0,
994
+ "coverage": 0.0,
995
+ "execute_generated": 0.0,
996
+ "fallback_rate": 1.0,
997
+ "hidden_chart_oracle_success": 1.0,
998
+ "hidden_chart_oracle_utility": 2.0,
999
+ "outcome_ptr": 0.0,
1000
+ "predicted_margin": -0.09430490853290777,
1001
+ "proposal_oracle_success": 1.0,
1002
+ "proposal_oracle_utility": 2.0,
1003
+ "selected_candidate_index": 1.0,
1004
+ "selected_success": 1.0,
1005
+ "selected_success_gain_over_base": 0.0,
1006
+ "selected_utility": 2.0,
1007
+ "selected_utility_gain_over_base": 0.0,
1008
+ "selector_regret": 0.0,
1009
+ "success_selector_gap": 0.0,
1010
+ "success_support_gap": 0.0,
1011
+ "support_gap": 0.0,
1012
+ "tau": 0.461885278016425
1013
+ },
1014
+ "1747178030/train2": {
1015
+ "base_success": 1.0,
1016
+ "base_utility": 2.0,
1017
+ "coverage": 0.0,
1018
+ "execute_generated": 0.0,
1019
+ "fallback_rate": 1.0,
1020
+ "hidden_chart_oracle_success": 1.0,
1021
+ "hidden_chart_oracle_utility": 2.0,
1022
+ "outcome_ptr": 0.0,
1023
+ "predicted_margin": -0.09430490853290777,
1024
+ "proposal_oracle_success": 1.0,
1025
+ "proposal_oracle_utility": 2.0,
1026
+ "selected_candidate_index": 9.0,
1027
+ "selected_success": 1.0,
1028
+ "selected_success_gain_over_base": 0.0,
1029
+ "selected_utility": 2.0,
1030
+ "selected_utility_gain_over_base": 0.0,
1031
+ "selector_regret": 0.0,
1032
+ "success_selector_gap": 0.0,
1033
+ "success_support_gap": 0.0,
1034
+ "support_gap": 0.0,
1035
+ "tau": 0.461885278016425
1036
+ },
1037
+ "1790188787/train0": {
1038
+ "base_success": 0.0,
1039
+ "base_utility": 0.12476746737957001,
1040
+ "coverage": 1.0,
1041
+ "execute_generated": 1.0,
1042
+ "fallback_rate": 0.0,
1043
+ "hidden_chart_oracle_success": 1.0,
1044
+ "hidden_chart_oracle_utility": 2.0,
1045
+ "outcome_ptr": 1.0,
1046
+ "predicted_margin": 1.928122443422234,
1047
+ "proposal_oracle_success": 1.0,
1048
+ "proposal_oracle_utility": 2.0,
1049
+ "selected_candidate_index": 4.0,
1050
+ "selected_success": 0.0,
1051
+ "selected_success_gain_over_base": 0.0,
1052
+ "selected_utility": 0.06047999858856201,
1053
+ "selected_utility_gain_over_base": -0.064287468791008,
1054
+ "selector_regret": 1.939520001411438,
1055
+ "success_selector_gap": 1.0,
1056
+ "success_support_gap": 0.0,
1057
+ "support_gap": 0.0,
1058
+ "tau": 0.461885278016425
1059
+ },
1060
+ "1790188787/train1": {
1061
+ "base_success": 0.0,
1062
+ "base_utility": 0.12476746737957001,
1063
+ "coverage": 1.0,
1064
+ "execute_generated": 1.0,
1065
+ "fallback_rate": 0.0,
1066
+ "hidden_chart_oracle_success": 1.0,
1067
+ "hidden_chart_oracle_utility": 2.0,
1068
+ "outcome_ptr": 1.0,
1069
+ "predicted_margin": 1.9487161705951446,
1070
+ "proposal_oracle_success": 1.0,
1071
+ "proposal_oracle_utility": 2.0,
1072
+ "selected_candidate_index": 3.0,
1073
+ "selected_success": 1.0,
1074
+ "selected_success_gain_over_base": 1.0,
1075
+ "selected_utility": 2.0,
1076
+ "selected_utility_gain_over_base": 1.87523253262043,
1077
+ "selector_regret": 0.0,
1078
+ "success_selector_gap": 0.0,
1079
+ "success_support_gap": 0.0,
1080
+ "support_gap": 0.0,
1081
+ "tau": 0.461885278016425
1082
+ },
1083
+ "1790188787/train2": {
1084
+ "base_success": 0.0,
1085
+ "base_utility": 0.12476746737957001,
1086
+ "coverage": 1.0,
1087
+ "execute_generated": 1.0,
1088
+ "fallback_rate": 0.0,
1089
+ "hidden_chart_oracle_success": 1.0,
1090
+ "hidden_chart_oracle_utility": 2.0,
1091
+ "outcome_ptr": 1.0,
1092
+ "predicted_margin": 1.9487161705951446,
1093
+ "proposal_oracle_success": 1.0,
1094
+ "proposal_oracle_utility": 2.0,
1095
+ "selected_candidate_index": 10.0,
1096
+ "selected_success": 0.0,
1097
+ "selected_success_gain_over_base": 0.0,
1098
+ "selected_utility": 0.10894416272640228,
1099
+ "selected_utility_gain_over_base": -0.015823304653167725,
1100
+ "selector_regret": 1.8910558372735977,
1101
+ "success_selector_gap": 1.0,
1102
+ "success_support_gap": 0.0,
1103
+ "support_gap": 0.0,
1104
+ "tau": 0.461885278016425
1105
+ },
1106
+ "1822210471/train0": {
1107
+ "base_success": 0.0,
1108
+ "base_utility": 0.07191530615091324,
1109
+ "coverage": 0.0,
1110
+ "execute_generated": 0.0,
1111
+ "fallback_rate": 1.0,
1112
+ "hidden_chart_oracle_success": 1.0,
1113
+ "hidden_chart_oracle_utility": 2.0,
1114
+ "outcome_ptr": 1.0,
1115
+ "predicted_margin": 0.3753131616482717,
1116
+ "proposal_oracle_success": 1.0,
1117
+ "proposal_oracle_utility": 2.0,
1118
+ "selected_candidate_index": 7.0,
1119
+ "selected_success": 0.0,
1120
+ "selected_success_gain_over_base": 0.0,
1121
+ "selected_utility": 0.07191530615091324,
1122
+ "selected_utility_gain_over_base": 0.0,
1123
+ "selector_regret": 1.9280846938490868,
1124
+ "success_selector_gap": 1.0,
1125
+ "success_support_gap": 0.0,
1126
+ "support_gap": 0.0,
1127
+ "tau": 0.461885278016425
1128
+ },
1129
+ "1822210471/train1": {
1130
+ "base_success": 0.0,
1131
+ "base_utility": 0.07191530615091324,
1132
+ "coverage": 0.0,
1133
+ "execute_generated": 0.0,
1134
+ "fallback_rate": 1.0,
1135
+ "hidden_chart_oracle_success": 1.0,
1136
+ "hidden_chart_oracle_utility": 2.0,
1137
+ "outcome_ptr": 1.0,
1138
+ "predicted_margin": 0.3462447613139206,
1139
+ "proposal_oracle_success": 1.0,
1140
+ "proposal_oracle_utility": 2.0,
1141
+ "selected_candidate_index": 0.0,
1142
+ "selected_success": 0.0,
1143
+ "selected_success_gain_over_base": 0.0,
1144
+ "selected_utility": 0.07191530615091324,
1145
+ "selected_utility_gain_over_base": 0.0,
1146
+ "selector_regret": 1.9280846938490868,
1147
+ "success_selector_gap": 1.0,
1148
+ "success_support_gap": 0.0,
1149
+ "support_gap": 0.0,
1150
+ "tau": 0.461885278016425
1151
+ },
1152
+ "1822210471/train2": {
1153
+ "base_success": 0.0,
1154
+ "base_utility": 0.07191530615091324,
1155
+ "coverage": 0.0,
1156
+ "execute_generated": 0.0,
1157
+ "fallback_rate": 1.0,
1158
+ "hidden_chart_oracle_success": 1.0,
1159
+ "hidden_chart_oracle_utility": 2.0,
1160
+ "outcome_ptr": 1.0,
1161
+ "predicted_margin": 0.3753131616482717,
1162
+ "proposal_oracle_success": 1.0,
1163
+ "proposal_oracle_utility": 2.0,
1164
+ "selected_candidate_index": 0.0,
1165
+ "selected_success": 0.0,
1166
+ "selected_success_gain_over_base": 0.0,
1167
+ "selected_utility": 0.07191530615091324,
1168
+ "selected_utility_gain_over_base": 0.0,
1169
+ "selector_regret": 1.9280846938490868,
1170
+ "success_selector_gap": 1.0,
1171
+ "success_support_gap": 0.0,
1172
+ "support_gap": 0.0,
1173
+ "tau": 0.461885278016425
1174
+ },
1175
+ "1845489013/train0": {
1176
+ "base_success": 1.0,
1177
+ "base_utility": 2.0,
1178
+ "coverage": 1.0,
1179
+ "execute_generated": 1.0,
1180
+ "fallback_rate": 0.0,
1181
+ "hidden_chart_oracle_success": 1.0,
1182
+ "hidden_chart_oracle_utility": 2.0,
1183
+ "outcome_ptr": 0.0,
1184
+ "predicted_margin": 0.46540267632512955,
1185
+ "proposal_oracle_success": 1.0,
1186
+ "proposal_oracle_utility": 2.0,
1187
+ "selected_candidate_index": 0.0,
1188
+ "selected_success": 1.0,
1189
+ "selected_success_gain_over_base": 0.0,
1190
+ "selected_utility": 2.0,
1191
+ "selected_utility_gain_over_base": 0.0,
1192
+ "selector_regret": 0.0,
1193
+ "success_selector_gap": 0.0,
1194
+ "success_support_gap": 0.0,
1195
+ "support_gap": 0.0,
1196
+ "tau": 0.461885278016425
1197
+ },
1198
+ "1845489013/train1": {
1199
+ "base_success": 1.0,
1200
+ "base_utility": 2.0,
1201
+ "coverage": 1.0,
1202
+ "execute_generated": 1.0,
1203
+ "fallback_rate": 0.0,
1204
+ "hidden_chart_oracle_success": 1.0,
1205
+ "hidden_chart_oracle_utility": 2.0,
1206
+ "outcome_ptr": 0.0,
1207
+ "predicted_margin": 0.46540267632512955,
1208
+ "proposal_oracle_success": 1.0,
1209
+ "proposal_oracle_utility": 2.0,
1210
+ "selected_candidate_index": 2.0,
1211
+ "selected_success": 1.0,
1212
+ "selected_success_gain_over_base": 0.0,
1213
+ "selected_utility": 2.0,
1214
+ "selected_utility_gain_over_base": 0.0,
1215
+ "selector_regret": 0.0,
1216
+ "success_selector_gap": 0.0,
1217
+ "success_support_gap": 0.0,
1218
+ "support_gap": 0.0,
1219
+ "tau": 0.461885278016425
1220
+ },
1221
+ "1845489013/train2": {
1222
+ "base_success": 1.0,
1223
+ "base_utility": 2.0,
1224
+ "coverage": 1.0,
1225
+ "execute_generated": 1.0,
1226
+ "fallback_rate": 0.0,
1227
+ "hidden_chart_oracle_success": 1.0,
1228
+ "hidden_chart_oracle_utility": 2.0,
1229
+ "outcome_ptr": 0.0,
1230
+ "predicted_margin": 0.46540267632512955,
1231
+ "proposal_oracle_success": 1.0,
1232
+ "proposal_oracle_utility": 2.0,
1233
+ "selected_candidate_index": 10.0,
1234
+ "selected_success": 0.0,
1235
+ "selected_success_gain_over_base": -1.0,
1236
+ "selected_utility": 0.08007363229990005,
1237
+ "selected_utility_gain_over_base": -1.9199263677001,
1238
+ "selector_regret": 1.9199263677001,
1239
+ "success_selector_gap": 1.0,
1240
+ "success_support_gap": 0.0,
1241
+ "support_gap": 0.0,
1242
+ "tau": 0.461885278016425
1243
+ },
1244
+ "1999022574/train0": {
1245
+ "base_success": 1.0,
1246
+ "base_utility": 2.0,
1247
+ "coverage": 0.0,
1248
+ "execute_generated": 0.0,
1249
+ "fallback_rate": 1.0,
1250
+ "hidden_chart_oracle_success": 1.0,
1251
+ "hidden_chart_oracle_utility": 2.0,
1252
+ "outcome_ptr": 0.0,
1253
+ "predicted_margin": 0.23966364189076328,
1254
+ "proposal_oracle_success": 1.0,
1255
+ "proposal_oracle_utility": 2.0,
1256
+ "selected_candidate_index": 5.0,
1257
+ "selected_success": 1.0,
1258
+ "selected_success_gain_over_base": 0.0,
1259
+ "selected_utility": 2.0,
1260
+ "selected_utility_gain_over_base": 0.0,
1261
+ "selector_regret": 0.0,
1262
+ "success_selector_gap": 0.0,
1263
+ "success_support_gap": 0.0,
1264
+ "support_gap": 0.0,
1265
+ "tau": 0.461885278016425
1266
+ },
1267
+ "1999022574/train1": {
1268
+ "base_success": 1.0,
1269
+ "base_utility": 2.0,
1270
+ "coverage": 0.0,
1271
+ "execute_generated": 0.0,
1272
+ "fallback_rate": 1.0,
1273
+ "hidden_chart_oracle_success": 1.0,
1274
+ "hidden_chart_oracle_utility": 2.0,
1275
+ "outcome_ptr": 0.0,
1276
+ "predicted_margin": 0.2690908362458581,
1277
+ "proposal_oracle_success": 1.0,
1278
+ "proposal_oracle_utility": 2.0,
1279
+ "selected_candidate_index": 8.0,
1280
+ "selected_success": 1.0,
1281
+ "selected_success_gain_over_base": 0.0,
1282
+ "selected_utility": 2.0,
1283
+ "selected_utility_gain_over_base": 0.0,
1284
+ "selector_regret": 0.0,
1285
+ "success_selector_gap": 0.0,
1286
+ "success_support_gap": 0.0,
1287
+ "support_gap": 0.0,
1288
+ "tau": 0.461885278016425
1289
+ },
1290
+ "1999022574/train2": {
1291
+ "base_success": 1.0,
1292
+ "base_utility": 2.0,
1293
+ "coverage": 0.0,
1294
+ "execute_generated": 0.0,
1295
+ "fallback_rate": 1.0,
1296
+ "hidden_chart_oracle_success": 1.0,
1297
+ "hidden_chart_oracle_utility": 2.0,
1298
+ "outcome_ptr": 0.0,
1299
+ "predicted_margin": 0.2690908362458581,
1300
+ "proposal_oracle_success": 1.0,
1301
+ "proposal_oracle_utility": 2.0,
1302
+ "selected_candidate_index": 7.0,
1303
+ "selected_success": 1.0,
1304
+ "selected_success_gain_over_base": 0.0,
1305
+ "selected_utility": 2.0,
1306
+ "selected_utility_gain_over_base": 0.0,
1307
+ "selector_regret": 0.0,
1308
+ "success_selector_gap": 0.0,
1309
+ "success_support_gap": 0.0,
1310
+ "support_gap": 0.0,
1311
+ "tau": 0.461885278016425
1312
+ },
1313
+ "2012211287/train0": {
1314
+ "base_success": 0.0,
1315
+ "base_utility": 0.7236053347587585,
1316
+ "coverage": 0.0,
1317
+ "execute_generated": 0.0,
1318
+ "fallback_rate": 1.0,
1319
+ "hidden_chart_oracle_success": 1.0,
1320
+ "hidden_chart_oracle_utility": 2.0,
1321
+ "outcome_ptr": 1.0,
1322
+ "predicted_margin": -0.10762542421499345,
1323
+ "proposal_oracle_success": 1.0,
1324
+ "proposal_oracle_utility": 2.0,
1325
+ "selected_candidate_index": 3.0,
1326
+ "selected_success": 0.0,
1327
+ "selected_success_gain_over_base": 0.0,
1328
+ "selected_utility": 0.7236053347587585,
1329
+ "selected_utility_gain_over_base": 0.0,
1330
+ "selector_regret": 1.2763946652412415,
1331
+ "success_selector_gap": 1.0,
1332
+ "success_support_gap": 0.0,
1333
+ "support_gap": 0.0,
1334
+ "tau": 0.461885278016425
1335
+ },
1336
+ "2012211287/train1": {
1337
+ "base_success": 0.0,
1338
+ "base_utility": 0.7236053347587585,
1339
+ "coverage": 0.0,
1340
+ "execute_generated": 0.0,
1341
+ "fallback_rate": 1.0,
1342
+ "hidden_chart_oracle_success": 1.0,
1343
+ "hidden_chart_oracle_utility": 2.0,
1344
+ "outcome_ptr": 1.0,
1345
+ "predicted_margin": -0.10762542421499345,
1346
+ "proposal_oracle_success": 1.0,
1347
+ "proposal_oracle_utility": 2.0,
1348
+ "selected_candidate_index": 7.0,
1349
+ "selected_success": 0.0,
1350
+ "selected_success_gain_over_base": 0.0,
1351
+ "selected_utility": 0.7236053347587585,
1352
+ "selected_utility_gain_over_base": 0.0,
1353
+ "selector_regret": 1.2763946652412415,
1354
+ "success_selector_gap": 1.0,
1355
+ "success_support_gap": 0.0,
1356
+ "support_gap": 0.0,
1357
+ "tau": 0.461885278016425
1358
+ },
1359
+ "2012211287/train2": {
1360
+ "base_success": 0.0,
1361
+ "base_utility": 0.7236053347587585,
1362
+ "coverage": 0.0,
1363
+ "execute_generated": 0.0,
1364
+ "fallback_rate": 1.0,
1365
+ "hidden_chart_oracle_success": 1.0,
1366
+ "hidden_chart_oracle_utility": 2.0,
1367
+ "outcome_ptr": 1.0,
1368
+ "predicted_margin": -0.10762542421499345,
1369
+ "proposal_oracle_success": 1.0,
1370
+ "proposal_oracle_utility": 2.0,
1371
+ "selected_candidate_index": 1.0,
1372
+ "selected_success": 0.0,
1373
+ "selected_success_gain_over_base": 0.0,
1374
+ "selected_utility": 0.7236053347587585,
1375
+ "selected_utility_gain_over_base": 0.0,
1376
+ "selector_regret": 1.2763946652412415,
1377
+ "success_selector_gap": 1.0,
1378
+ "success_support_gap": 0.0,
1379
+ "support_gap": 0.0,
1380
+ "tau": 0.461885278016425
1381
+ },
1382
+ "2038594191/train0": {
1383
+ "base_success": 1.0,
1384
+ "base_utility": 2.0,
1385
+ "coverage": 0.0,
1386
+ "execute_generated": 0.0,
1387
+ "fallback_rate": 1.0,
1388
+ "hidden_chart_oracle_success": 1.0,
1389
+ "hidden_chart_oracle_utility": 2.0,
1390
+ "outcome_ptr": 0.0,
1391
+ "predicted_margin": 0.4320928427625122,
1392
+ "proposal_oracle_success": 1.0,
1393
+ "proposal_oracle_utility": 2.0,
1394
+ "selected_candidate_index": 3.0,
1395
+ "selected_success": 1.0,
1396
+ "selected_success_gain_over_base": 0.0,
1397
+ "selected_utility": 2.0,
1398
+ "selected_utility_gain_over_base": 0.0,
1399
+ "selector_regret": 0.0,
1400
+ "success_selector_gap": 0.0,
1401
+ "success_support_gap": 0.0,
1402
+ "support_gap": 0.0,
1403
+ "tau": 0.461885278016425
1404
+ },
1405
+ "2038594191/train1": {
1406
+ "base_success": 1.0,
1407
+ "base_utility": 2.0,
1408
+ "coverage": 0.0,
1409
+ "execute_generated": 0.0,
1410
+ "fallback_rate": 1.0,
1411
+ "hidden_chart_oracle_success": 1.0,
1412
+ "hidden_chart_oracle_utility": 2.0,
1413
+ "outcome_ptr": 0.0,
1414
+ "predicted_margin": 0.4320928427625122,
1415
+ "proposal_oracle_success": 1.0,
1416
+ "proposal_oracle_utility": 2.0,
1417
+ "selected_candidate_index": 0.0,
1418
+ "selected_success": 1.0,
1419
+ "selected_success_gain_over_base": 0.0,
1420
+ "selected_utility": 2.0,
1421
+ "selected_utility_gain_over_base": 0.0,
1422
+ "selector_regret": 0.0,
1423
+ "success_selector_gap": 0.0,
1424
+ "success_support_gap": 0.0,
1425
+ "support_gap": 0.0,
1426
+ "tau": 0.461885278016425
1427
+ },
1428
+ "2038594191/train2": {
1429
+ "base_success": 1.0,
1430
+ "base_utility": 2.0,
1431
+ "coverage": 0.0,
1432
+ "execute_generated": 0.0,
1433
+ "fallback_rate": 1.0,
1434
+ "hidden_chart_oracle_success": 1.0,
1435
+ "hidden_chart_oracle_utility": 2.0,
1436
+ "outcome_ptr": 0.0,
1437
+ "predicted_margin": 0.4320928427625122,
1438
+ "proposal_oracle_success": 1.0,
1439
+ "proposal_oracle_utility": 2.0,
1440
+ "selected_candidate_index": 1.0,
1441
+ "selected_success": 1.0,
1442
+ "selected_success_gain_over_base": 0.0,
1443
+ "selected_utility": 2.0,
1444
+ "selected_utility_gain_over_base": 0.0,
1445
+ "selector_regret": 0.0,
1446
+ "success_selector_gap": 0.0,
1447
+ "success_support_gap": 0.0,
1448
+ "support_gap": 0.0,
1449
+ "tau": 0.461885278016425
1450
+ },
1451
+ "2054899126/train0": {
1452
+ "base_success": 1.0,
1453
+ "base_utility": 2.0,
1454
+ "coverage": 0.5,
1455
+ "execute_generated": 0.5,
1456
+ "fallback_rate": 0.5,
1457
+ "hidden_chart_oracle_success": 1.0,
1458
+ "hidden_chart_oracle_utility": 2.0,
1459
+ "outcome_ptr": 0.0,
1460
+ "predicted_margin": 0.8102161823738483,
1461
+ "proposal_oracle_success": 0.5,
1462
+ "proposal_oracle_utility": 1.077283151447773,
1463
+ "selected_candidate_index": 6.0,
1464
+ "selected_success": 0.5,
1465
+ "selected_success_gain_over_base": -0.5,
1466
+ "selected_utility": 1.052448581904173,
1467
+ "selected_utility_gain_over_base": -0.9475514180958271,
1468
+ "selector_regret": 0.024834569543600082,
1469
+ "success_selector_gap": 0.0,
1470
+ "success_support_gap": 0.5,
1471
+ "support_gap": 0.922716848552227,
1472
+ "tau": 0.461885278016425
1473
+ },
1474
+ "2054899126/train1": {
1475
+ "base_success": 1.0,
1476
+ "base_utility": 2.0,
1477
+ "coverage": 0.5,
1478
+ "execute_generated": 0.5,
1479
+ "fallback_rate": 0.5,
1480
+ "hidden_chart_oracle_success": 1.0,
1481
+ "hidden_chart_oracle_utility": 2.0,
1482
+ "outcome_ptr": 0.0,
1483
+ "predicted_margin": 0.7623409522984086,
1484
+ "proposal_oracle_success": 1.0,
1485
+ "proposal_oracle_utility": 2.0,
1486
+ "selected_candidate_index": 0.0,
1487
+ "selected_success": 0.5,
1488
+ "selected_success_gain_over_base": -0.5,
1489
+ "selected_utility": 1.0664558708667755,
1490
+ "selected_utility_gain_over_base": -0.9335441291332245,
1491
+ "selector_regret": 0.9335441291332245,
1492
+ "success_selector_gap": 0.5,
1493
+ "success_support_gap": 0.0,
1494
+ "support_gap": 0.0,
1495
+ "tau": 0.461885278016425
1496
+ },
1497
+ "2054899126/train2": {
1498
+ "base_success": 1.0,
1499
+ "base_utility": 2.0,
1500
+ "coverage": 0.5,
1501
+ "execute_generated": 0.5,
1502
+ "fallback_rate": 0.5,
1503
+ "hidden_chart_oracle_success": 1.0,
1504
+ "hidden_chart_oracle_utility": 2.0,
1505
+ "outcome_ptr": 0.0,
1506
+ "predicted_margin": 0.5871931326735697,
1507
+ "proposal_oracle_success": 1.0,
1508
+ "proposal_oracle_utility": 2.0,
1509
+ "selected_candidate_index": 2.5,
1510
+ "selected_success": 0.5,
1511
+ "selected_success_gain_over_base": -0.5,
1512
+ "selected_utility": 1.0387911088764668,
1513
+ "selected_utility_gain_over_base": -0.9612088911235332,
1514
+ "selector_regret": 0.9612088911235332,
1515
+ "success_selector_gap": 0.5,
1516
+ "success_support_gap": 0.0,
1517
+ "support_gap": 0.0,
1518
+ "tau": 0.461885278016425
1519
+ },
1520
+ "2061635888/train0": {
1521
+ "base_success": 0.0,
1522
+ "base_utility": 0.05225050449371338,
1523
+ "coverage": 0.0,
1524
+ "execute_generated": 0.0,
1525
+ "fallback_rate": 1.0,
1526
+ "hidden_chart_oracle_success": 1.0,
1527
+ "hidden_chart_oracle_utility": 2.0,
1528
+ "outcome_ptr": 1.0,
1529
+ "predicted_margin": -0.01765980271334038,
1530
+ "proposal_oracle_success": 0.0,
1531
+ "proposal_oracle_utility": 0.09799930453300476,
1532
+ "selected_candidate_index": 2.0,
1533
+ "selected_success": 0.0,
1534
+ "selected_success_gain_over_base": 0.0,
1535
+ "selected_utility": 0.05225050449371338,
1536
+ "selected_utility_gain_over_base": 0.0,
1537
+ "selector_regret": 0.04574880003929138,
1538
+ "success_selector_gap": 0.0,
1539
+ "success_support_gap": 1.0,
1540
+ "support_gap": 1.9020006954669952,
1541
+ "tau": 0.461885278016425
1542
+ },
1543
+ "2061635888/train1": {
1544
+ "base_success": 0.0,
1545
+ "base_utility": 0.05225050449371338,
1546
+ "coverage": 0.0,
1547
+ "execute_generated": 0.0,
1548
+ "fallback_rate": 1.0,
1549
+ "hidden_chart_oracle_success": 1.0,
1550
+ "hidden_chart_oracle_utility": 2.0,
1551
+ "outcome_ptr": 1.0,
1552
+ "predicted_margin": 0.01069560631150106,
1553
+ "proposal_oracle_success": 0.0,
1554
+ "proposal_oracle_utility": 0.07580336928367615,
1555
+ "selected_candidate_index": 0.0,
1556
+ "selected_success": 0.0,
1557
+ "selected_success_gain_over_base": 0.0,
1558
+ "selected_utility": 0.05225050449371338,
1559
+ "selected_utility_gain_over_base": 0.0,
1560
+ "selector_regret": 0.02355286478996277,
1561
+ "success_selector_gap": 0.0,
1562
+ "success_support_gap": 1.0,
1563
+ "support_gap": 1.9241966307163239,
1564
+ "tau": 0.461885278016425
1565
+ },
1566
+ "2061635888/train2": {
1567
+ "base_success": 0.0,
1568
+ "base_utility": 0.05225050449371338,
1569
+ "coverage": 0.0,
1570
+ "execute_generated": 0.0,
1571
+ "fallback_rate": 1.0,
1572
+ "hidden_chart_oracle_success": 1.0,
1573
+ "hidden_chart_oracle_utility": 2.0,
1574
+ "outcome_ptr": 1.0,
1575
+ "predicted_margin": 0.01069560631150106,
1576
+ "proposal_oracle_success": 0.0,
1577
+ "proposal_oracle_utility": 0.11105132102966309,
1578
+ "selected_candidate_index": 2.0,
1579
+ "selected_success": 0.0,
1580
+ "selected_success_gain_over_base": 0.0,
1581
+ "selected_utility": 0.05225050449371338,
1582
+ "selected_utility_gain_over_base": 0.0,
1583
+ "selector_regret": 0.05880081653594971,
1584
+ "success_selector_gap": 0.0,
1585
+ "success_support_gap": 1.0,
1586
+ "support_gap": 1.888948678970337,
1587
+ "tau": 0.461885278016425
1588
+ },
1589
+ "211245842/train0": {
1590
+ "base_success": 0.0,
1591
+ "base_utility": 0.19090841710567474,
1592
+ "coverage": 1.0,
1593
+ "execute_generated": 1.0,
1594
+ "fallback_rate": 0.0,
1595
+ "hidden_chart_oracle_success": 1.0,
1596
+ "hidden_chart_oracle_utility": 2.0,
1597
+ "outcome_ptr": 1.0,
1598
+ "predicted_margin": 0.6258640055379121,
1599
+ "proposal_oracle_success": 1.0,
1600
+ "proposal_oracle_utility": 2.0,
1601
+ "selected_candidate_index": 4.0,
1602
+ "selected_success": 0.0,
1603
+ "selected_success_gain_over_base": 0.0,
1604
+ "selected_utility": 0.1094239205121994,
1605
+ "selected_utility_gain_over_base": -0.08148449659347534,
1606
+ "selector_regret": 1.8905760794878006,
1607
+ "success_selector_gap": 1.0,
1608
+ "success_support_gap": 0.0,
1609
+ "support_gap": 0.0,
1610
+ "tau": 0.461885278016425
1611
+ },
1612
+ "211245842/train1": {
1613
+ "base_success": 0.0,
1614
+ "base_utility": 0.19090841710567474,
1615
+ "coverage": 1.0,
1616
+ "execute_generated": 1.0,
1617
+ "fallback_rate": 0.0,
1618
+ "hidden_chart_oracle_success": 1.0,
1619
+ "hidden_chart_oracle_utility": 2.0,
1620
+ "outcome_ptr": 1.0,
1621
+ "predicted_margin": 0.5604320834866071,
1622
+ "proposal_oracle_success": 0.0,
1623
+ "proposal_oracle_utility": 0.21295717358589172,
1624
+ "selected_candidate_index": 2.0,
1625
+ "selected_success": 0.0,
1626
+ "selected_success_gain_over_base": 0.0,
1627
+ "selected_utility": 0.18476036190986633,
1628
+ "selected_utility_gain_over_base": -0.006148055195808411,
1629
+ "selector_regret": 0.02819681167602539,
1630
+ "success_selector_gap": 0.0,
1631
+ "success_support_gap": 1.0,
1632
+ "support_gap": 1.7870428264141083,
1633
+ "tau": 0.461885278016425
1634
+ },
1635
+ "211245842/train2": {
1636
+ "base_success": 0.0,
1637
+ "base_utility": 0.19090841710567474,
1638
+ "coverage": 1.0,
1639
+ "execute_generated": 1.0,
1640
+ "fallback_rate": 0.0,
1641
+ "hidden_chart_oracle_success": 1.0,
1642
+ "hidden_chart_oracle_utility": 2.0,
1643
+ "outcome_ptr": 1.0,
1644
+ "predicted_margin": 0.6258640055379121,
1645
+ "proposal_oracle_success": 0.0,
1646
+ "proposal_oracle_utility": 0.23388276994228363,
1647
+ "selected_candidate_index": 10.0,
1648
+ "selected_success": 0.0,
1649
+ "selected_success_gain_over_base": 0.0,
1650
+ "selected_utility": 0.12799203395843506,
1651
+ "selected_utility_gain_over_base": -0.06291638314723969,
1652
+ "selector_regret": 0.10589073598384857,
1653
+ "success_selector_gap": 0.0,
1654
+ "success_support_gap": 1.0,
1655
+ "support_gap": 1.7661172300577164,
1656
+ "tau": 0.461885278016425
1657
+ },
1658
+ "278008524/train0": {
1659
+ "base_success": 0.0,
1660
+ "base_utility": 0.042734989896416664,
1661
+ "coverage": 0.0,
1662
+ "execute_generated": 0.0,
1663
+ "fallback_rate": 1.0,
1664
+ "hidden_chart_oracle_success": 0.5,
1665
+ "hidden_chart_oracle_utility": 1.1139086410403252,
1666
+ "outcome_ptr": 1.0,
1667
+ "predicted_margin": 0.18325703686238062,
1668
+ "proposal_oracle_success": 0.5,
1669
+ "proposal_oracle_utility": 1.0733563229441643,
1670
+ "selected_candidate_index": 0.0,
1671
+ "selected_success": 0.0,
1672
+ "selected_success_gain_over_base": 0.0,
1673
+ "selected_utility": 0.042734989896416664,
1674
+ "selected_utility_gain_over_base": 0.0,
1675
+ "selector_regret": 1.0306213330477476,
1676
+ "success_selector_gap": 0.5,
1677
+ "success_support_gap": 0.0,
1678
+ "support_gap": 0.04055231809616089,
1679
+ "tau": 0.461885278016425
1680
+ },
1681
+ "278008524/train1": {
1682
+ "base_success": 0.0,
1683
+ "base_utility": 0.042734989896416664,
1684
+ "coverage": 0.0,
1685
+ "execute_generated": 0.0,
1686
+ "fallback_rate": 1.0,
1687
+ "hidden_chart_oracle_success": 0.5,
1688
+ "hidden_chart_oracle_utility": 1.1139086410403252,
1689
+ "outcome_ptr": 1.0,
1690
+ "predicted_margin": 0.18325703686238062,
1691
+ "proposal_oracle_success": 0.5,
1692
+ "proposal_oracle_utility": 1.0376745946705341,
1693
+ "selected_candidate_index": 4.5,
1694
+ "selected_success": 0.0,
1695
+ "selected_success_gain_over_base": 0.0,
1696
+ "selected_utility": 0.042734989896416664,
1697
+ "selected_utility_gain_over_base": 0.0,
1698
+ "selector_regret": 0.9949396047741175,
1699
+ "success_selector_gap": 0.5,
1700
+ "success_support_gap": 0.0,
1701
+ "support_gap": 0.07623404636979103,
1702
+ "tau": 0.461885278016425
1703
+ },
1704
+ "278008524/train2": {
1705
+ "base_success": 0.0,
1706
+ "base_utility": 0.042734989896416664,
1707
+ "coverage": 0.0,
1708
+ "execute_generated": 0.0,
1709
+ "fallback_rate": 1.0,
1710
+ "hidden_chart_oracle_success": 0.5,
1711
+ "hidden_chart_oracle_utility": 1.1139086410403252,
1712
+ "outcome_ptr": 1.0,
1713
+ "predicted_margin": 0.18325703686238062,
1714
+ "proposal_oracle_success": 0.5,
1715
+ "proposal_oracle_utility": 1.1359489113092422,
1716
+ "selected_candidate_index": 0.5,
1717
+ "selected_success": 0.0,
1718
+ "selected_success_gain_over_base": 0.0,
1719
+ "selected_utility": 0.042734989896416664,
1720
+ "selected_utility_gain_over_base": 0.0,
1721
+ "selector_regret": 1.0932139214128256,
1722
+ "success_selector_gap": 0.5,
1723
+ "success_support_gap": 0.0,
1724
+ "support_gap": 0.0,
1725
+ "tau": 0.461885278016425
1726
+ },
1727
+ "307068879/train0": {
1728
+ "base_success": 0.0,
1729
+ "base_utility": 0.6000905632972717,
1730
+ "coverage": 0.0,
1731
+ "execute_generated": 0.0,
1732
+ "fallback_rate": 1.0,
1733
+ "hidden_chart_oracle_success": 1.0,
1734
+ "hidden_chart_oracle_utility": 2.0,
1735
+ "outcome_ptr": 0.0,
1736
+ "predicted_margin": 0.03189693906883312,
1737
+ "proposal_oracle_success": 0.0,
1738
+ "proposal_oracle_utility": 0.2115575522184372,
1739
+ "selected_candidate_index": 4.0,
1740
+ "selected_success": 0.0,
1741
+ "selected_success_gain_over_base": 0.0,
1742
+ "selected_utility": 0.6000905632972717,
1743
+ "selected_utility_gain_over_base": 0.0,
1744
+ "selector_regret": 0.0,
1745
+ "success_selector_gap": 0.0,
1746
+ "success_support_gap": 1.0,
1747
+ "support_gap": 1.7884424477815628,
1748
+ "tau": 0.461885278016425
1749
+ },
1750
+ "307068879/train1": {
1751
+ "base_success": 0.0,
1752
+ "base_utility": 0.6000905632972717,
1753
+ "coverage": 0.0,
1754
+ "execute_generated": 0.0,
1755
+ "fallback_rate": 1.0,
1756
+ "hidden_chart_oracle_success": 1.0,
1757
+ "hidden_chart_oracle_utility": 2.0,
1758
+ "outcome_ptr": 0.0,
1759
+ "predicted_margin": 0.024889983439393283,
1760
+ "proposal_oracle_success": 0.0,
1761
+ "proposal_oracle_utility": 0.5876073241233826,
1762
+ "selected_candidate_index": 0.0,
1763
+ "selected_success": 0.0,
1764
+ "selected_success_gain_over_base": 0.0,
1765
+ "selected_utility": 0.6000905632972717,
1766
+ "selected_utility_gain_over_base": 0.0,
1767
+ "selector_regret": 0.0,
1768
+ "success_selector_gap": 0.0,
1769
+ "success_support_gap": 1.0,
1770
+ "support_gap": 1.4123926758766174,
1771
+ "tau": 0.461885278016425
1772
+ },
1773
+ "307068879/train2": {
1774
+ "base_success": 0.0,
1775
+ "base_utility": 0.6000905632972717,
1776
+ "coverage": 0.0,
1777
+ "execute_generated": 0.0,
1778
+ "fallback_rate": 1.0,
1779
+ "hidden_chart_oracle_success": 1.0,
1780
+ "hidden_chart_oracle_utility": 2.0,
1781
+ "outcome_ptr": 0.0,
1782
+ "predicted_margin": 0.045483710612303836,
1783
+ "proposal_oracle_success": 0.0,
1784
+ "proposal_oracle_utility": 0.2302134484052658,
1785
+ "selected_candidate_index": 4.0,
1786
+ "selected_success": 0.0,
1787
+ "selected_success_gain_over_base": 0.0,
1788
+ "selected_utility": 0.6000905632972717,
1789
+ "selected_utility_gain_over_base": 0.0,
1790
+ "selector_regret": 0.0,
1791
+ "success_selector_gap": 0.0,
1792
+ "success_support_gap": 1.0,
1793
+ "support_gap": 1.7697865515947342,
1794
+ "tau": 0.461885278016425
1795
+ },
1796
+ "412102563/train0": {
1797
+ "base_success": 1.0,
1798
+ "base_utility": 2.0,
1799
+ "coverage": 0.0,
1800
+ "execute_generated": 0.0,
1801
+ "fallback_rate": 1.0,
1802
+ "hidden_chart_oracle_success": 1.0,
1803
+ "hidden_chart_oracle_utility": 2.0,
1804
+ "outcome_ptr": 0.0,
1805
+ "predicted_margin": 0.2296842211404253,
1806
+ "proposal_oracle_success": 1.0,
1807
+ "proposal_oracle_utility": 2.0,
1808
+ "selected_candidate_index": 2.0,
1809
+ "selected_success": 1.0,
1810
+ "selected_success_gain_over_base": 0.0,
1811
+ "selected_utility": 2.0,
1812
+ "selected_utility_gain_over_base": 0.0,
1813
+ "selector_regret": 0.0,
1814
+ "success_selector_gap": 0.0,
1815
+ "success_support_gap": 0.0,
1816
+ "support_gap": 0.0,
1817
+ "tau": 0.461885278016425
1818
+ },
1819
+ "412102563/train1": {
1820
+ "base_success": 1.0,
1821
+ "base_utility": 2.0,
1822
+ "coverage": 0.0,
1823
+ "execute_generated": 0.0,
1824
+ "fallback_rate": 1.0,
1825
+ "hidden_chart_oracle_success": 1.0,
1826
+ "hidden_chart_oracle_utility": 2.0,
1827
+ "outcome_ptr": 0.0,
1828
+ "predicted_margin": 0.25659152366583776,
1829
+ "proposal_oracle_success": 1.0,
1830
+ "proposal_oracle_utility": 2.0,
1831
+ "selected_candidate_index": 10.0,
1832
+ "selected_success": 1.0,
1833
+ "selected_success_gain_over_base": 0.0,
1834
+ "selected_utility": 2.0,
1835
+ "selected_utility_gain_over_base": 0.0,
1836
+ "selector_regret": 0.0,
1837
+ "success_selector_gap": 0.0,
1838
+ "success_support_gap": 0.0,
1839
+ "support_gap": 0.0,
1840
+ "tau": 0.461885278016425
1841
+ },
1842
+ "412102563/train2": {
1843
+ "base_success": 1.0,
1844
+ "base_utility": 2.0,
1845
+ "coverage": 0.0,
1846
+ "execute_generated": 0.0,
1847
+ "fallback_rate": 1.0,
1848
+ "hidden_chart_oracle_success": 1.0,
1849
+ "hidden_chart_oracle_utility": 2.0,
1850
+ "outcome_ptr": 0.0,
1851
+ "predicted_margin": 0.2296842211404253,
1852
+ "proposal_oracle_success": 1.0,
1853
+ "proposal_oracle_utility": 2.0,
1854
+ "selected_candidate_index": 5.0,
1855
+ "selected_success": 1.0,
1856
+ "selected_success_gain_over_base": 0.0,
1857
+ "selected_utility": 2.0,
1858
+ "selected_utility_gain_over_base": 0.0,
1859
+ "selector_regret": 0.0,
1860
+ "success_selector_gap": 0.0,
1861
+ "success_support_gap": 0.0,
1862
+ "support_gap": 0.0,
1863
+ "tau": 0.461885278016425
1864
+ },
1865
+ "440577711/train0": {
1866
+ "base_success": 0.0,
1867
+ "base_utility": 0.036152780055999756,
1868
+ "coverage": 1.0,
1869
+ "execute_generated": 1.0,
1870
+ "fallback_rate": 0.0,
1871
+ "hidden_chart_oracle_success": 1.0,
1872
+ "hidden_chart_oracle_utility": 2.0,
1873
+ "outcome_ptr": 1.0,
1874
+ "predicted_margin": 0.7208522520964407,
1875
+ "proposal_oracle_success": 1.0,
1876
+ "proposal_oracle_utility": 2.0,
1877
+ "selected_candidate_index": 0.0,
1878
+ "selected_success": 1.0,
1879
+ "selected_success_gain_over_base": 1.0,
1880
+ "selected_utility": 2.0,
1881
+ "selected_utility_gain_over_base": 1.9638472199440002,
1882
+ "selector_regret": 0.0,
1883
+ "success_selector_gap": 0.0,
1884
+ "success_support_gap": 0.0,
1885
+ "support_gap": 0.0,
1886
+ "tau": 0.461885278016425
1887
+ },
1888
+ "440577711/train1": {
1889
+ "base_success": 0.0,
1890
+ "base_utility": 0.036152780055999756,
1891
+ "coverage": 1.0,
1892
+ "execute_generated": 1.0,
1893
+ "fallback_rate": 0.0,
1894
+ "hidden_chart_oracle_success": 1.0,
1895
+ "hidden_chart_oracle_utility": 2.0,
1896
+ "outcome_ptr": 1.0,
1897
+ "predicted_margin": 0.8035469324826213,
1898
+ "proposal_oracle_success": 1.0,
1899
+ "proposal_oracle_utility": 2.0,
1900
+ "selected_candidate_index": 3.0,
1901
+ "selected_success": 1.0,
1902
+ "selected_success_gain_over_base": 1.0,
1903
+ "selected_utility": 2.0,
1904
+ "selected_utility_gain_over_base": 1.9638472199440002,
1905
+ "selector_regret": 0.0,
1906
+ "success_selector_gap": 0.0,
1907
+ "success_support_gap": 0.0,
1908
+ "support_gap": 0.0,
1909
+ "tau": 0.461885278016425
1910
+ },
1911
+ "440577711/train2": {
1912
+ "base_success": 0.0,
1913
+ "base_utility": 0.036152780055999756,
1914
+ "coverage": 1.0,
1915
+ "execute_generated": 1.0,
1916
+ "fallback_rate": 0.0,
1917
+ "hidden_chart_oracle_success": 1.0,
1918
+ "hidden_chart_oracle_utility": 2.0,
1919
+ "outcome_ptr": 1.0,
1920
+ "predicted_margin": 0.7208522520964407,
1921
+ "proposal_oracle_success": 1.0,
1922
+ "proposal_oracle_utility": 2.0,
1923
+ "selected_candidate_index": 1.0,
1924
+ "selected_success": 1.0,
1925
+ "selected_success_gain_over_base": 1.0,
1926
+ "selected_utility": 2.0,
1927
+ "selected_utility_gain_over_base": 1.9638472199440002,
1928
+ "selector_regret": 0.0,
1929
+ "success_selector_gap": 0.0,
1930
+ "success_support_gap": 0.0,
1931
+ "support_gap": 0.0,
1932
+ "tau": 0.461885278016425
1933
+ },
1934
+ "441046872/train0": {
1935
+ "base_success": 0.0,
1936
+ "base_utility": 0.6999551653862,
1937
+ "coverage": 0.0,
1938
+ "execute_generated": 0.0,
1939
+ "fallback_rate": 1.0,
1940
+ "hidden_chart_oracle_success": 1.0,
1941
+ "hidden_chart_oracle_utility": 2.0,
1942
+ "outcome_ptr": 1.0,
1943
+ "predicted_margin": 0.01069560631150106,
1944
+ "proposal_oracle_success": 1.0,
1945
+ "proposal_oracle_utility": 2.0,
1946
+ "selected_candidate_index": 1.0,
1947
+ "selected_success": 0.0,
1948
+ "selected_success_gain_over_base": 0.0,
1949
+ "selected_utility": 0.6999551653862,
1950
+ "selected_utility_gain_over_base": 0.0,
1951
+ "selector_regret": 1.3000448346138,
1952
+ "success_selector_gap": 1.0,
1953
+ "success_support_gap": 0.0,
1954
+ "support_gap": 0.0,
1955
+ "tau": 0.461885278016425
1956
+ },
1957
+ "441046872/train1": {
1958
+ "base_success": 0.0,
1959
+ "base_utility": 0.6999551653862,
1960
+ "coverage": 0.0,
1961
+ "execute_generated": 0.0,
1962
+ "fallback_rate": 1.0,
1963
+ "hidden_chart_oracle_success": 1.0,
1964
+ "hidden_chart_oracle_utility": 2.0,
1965
+ "outcome_ptr": 1.0,
1966
+ "predicted_margin": 0.01069560631150106,
1967
+ "proposal_oracle_success": 0.0,
1968
+ "proposal_oracle_utility": 0.7233588695526123,
1969
+ "selected_candidate_index": 1.0,
1970
+ "selected_success": 0.0,
1971
+ "selected_success_gain_over_base": 0.0,
1972
+ "selected_utility": 0.6999551653862,
1973
+ "selected_utility_gain_over_base": 0.0,
1974
+ "selector_regret": 0.023403704166412354,
1975
+ "success_selector_gap": 0.0,
1976
+ "success_support_gap": 1.0,
1977
+ "support_gap": 1.2766411304473877,
1978
+ "tau": 0.461885278016425
1979
+ },
1980
+ "441046872/train2": {
1981
+ "base_success": 0.0,
1982
+ "base_utility": 0.6999551653862,
1983
+ "coverage": 0.0,
1984
+ "execute_generated": 0.0,
1985
+ "fallback_rate": 1.0,
1986
+ "hidden_chart_oracle_success": 1.0,
1987
+ "hidden_chart_oracle_utility": 2.0,
1988
+ "outcome_ptr": 1.0,
1989
+ "predicted_margin": 0.01069560631150106,
1990
+ "proposal_oracle_success": 0.0,
1991
+ "proposal_oracle_utility": 0.7191348671913147,
1992
+ "selected_candidate_index": 0.0,
1993
+ "selected_success": 0.0,
1994
+ "selected_success_gain_over_base": 0.0,
1995
+ "selected_utility": 0.6999551653862,
1996
+ "selected_utility_gain_over_base": 0.0,
1997
+ "selector_regret": 0.019179701805114746,
1998
+ "success_selector_gap": 0.0,
1999
+ "success_support_gap": 1.0,
2000
+ "support_gap": 1.2808651328086853,
2001
+ "tau": 0.461885278016425
2002
+ },
2003
+ "509430392/train0": {
2004
+ "base_success": 0.0,
2005
+ "base_utility": 2.2351741790771484e-07,
2006
+ "coverage": 1.0,
2007
+ "execute_generated": 1.0,
2008
+ "fallback_rate": 0.0,
2009
+ "hidden_chart_oracle_success": 1.0,
2010
+ "hidden_chart_oracle_utility": 2.0,
2011
+ "outcome_ptr": 1.0,
2012
+ "predicted_margin": 1.425471802912922,
2013
+ "proposal_oracle_success": 1.0,
2014
+ "proposal_oracle_utility": 2.0,
2015
+ "selected_candidate_index": 14.0,
2016
+ "selected_success": 0.0,
2017
+ "selected_success_gain_over_base": 0.0,
2018
+ "selected_utility": 0.12097799777984619,
2019
+ "selected_utility_gain_over_base": 0.12097777426242828,
2020
+ "selector_regret": 1.8790220022201538,
2021
+ "success_selector_gap": 1.0,
2022
+ "success_support_gap": 0.0,
2023
+ "support_gap": 0.0,
2024
+ "tau": 0.461885278016425
2025
+ },
2026
+ "509430392/train1": {
2027
+ "base_success": 0.0,
2028
+ "base_utility": 2.2351741790771484e-07,
2029
+ "coverage": 1.0,
2030
+ "execute_generated": 1.0,
2031
+ "fallback_rate": 0.0,
2032
+ "hidden_chart_oracle_success": 1.0,
2033
+ "hidden_chart_oracle_utility": 2.0,
2034
+ "outcome_ptr": 1.0,
2035
+ "predicted_margin": 1.4513022033896152,
2036
+ "proposal_oracle_success": 1.0,
2037
+ "proposal_oracle_utility": 2.0,
2038
+ "selected_candidate_index": 2.0,
2039
+ "selected_success": 1.0,
2040
+ "selected_success_gain_over_base": 1.0,
2041
+ "selected_utility": 2.0,
2042
+ "selected_utility_gain_over_base": 1.999999776482582,
2043
+ "selector_regret": 0.0,
2044
+ "success_selector_gap": 0.0,
2045
+ "success_support_gap": 0.0,
2046
+ "support_gap": 0.0,
2047
+ "tau": 0.461885278016425
2048
+ },
2049
+ "509430392/train2": {
2050
+ "base_success": 0.0,
2051
+ "base_utility": 2.2351741790771484e-07,
2052
+ "coverage": 1.0,
2053
+ "execute_generated": 1.0,
2054
+ "fallback_rate": 0.0,
2055
+ "hidden_chart_oracle_success": 1.0,
2056
+ "hidden_chart_oracle_utility": 2.0,
2057
+ "outcome_ptr": 1.0,
2058
+ "predicted_margin": 1.425471802912922,
2059
+ "proposal_oracle_success": 0.0,
2060
+ "proposal_oracle_utility": 5.424022674560547e-06,
2061
+ "selected_candidate_index": 1.0,
2062
+ "selected_success": 0.0,
2063
+ "selected_success_gain_over_base": 0.0,
2064
+ "selected_utility": 2.682209014892578e-07,
2065
+ "selected_utility_gain_over_base": 4.470348358154297e-08,
2066
+ "selector_regret": 5.155801773071289e-06,
2067
+ "success_selector_gap": 0.0,
2068
+ "success_support_gap": 1.0,
2069
+ "support_gap": 1.9999945759773254,
2070
+ "tau": 0.461885278016425
2071
+ },
2072
+ "585904130/train0": {
2073
+ "base_success": 0.0,
2074
+ "base_utility": 0.18171554803848267,
2075
+ "coverage": 1.0,
2076
+ "execute_generated": 1.0,
2077
+ "fallback_rate": 0.0,
2078
+ "hidden_chart_oracle_success": 1.0,
2079
+ "hidden_chart_oracle_utility": 2.0,
2080
+ "outcome_ptr": 1.0,
2081
+ "predicted_margin": 0.783211815160911,
2082
+ "proposal_oracle_success": 1.0,
2083
+ "proposal_oracle_utility": 2.0,
2084
+ "selected_candidate_index": 4.0,
2085
+ "selected_success": 0.0,
2086
+ "selected_success_gain_over_base": 0.0,
2087
+ "selected_utility": 0.09919407218694687,
2088
+ "selected_utility_gain_over_base": -0.0825214758515358,
2089
+ "selector_regret": 1.9008059278130531,
2090
+ "success_selector_gap": 1.0,
2091
+ "success_support_gap": 0.0,
2092
+ "support_gap": 0.0,
2093
+ "tau": 0.461885278016425
2094
+ },
2095
+ "585904130/train1": {
2096
+ "base_success": 0.0,
2097
+ "base_utility": 0.18171554803848267,
2098
+ "coverage": 1.0,
2099
+ "execute_generated": 1.0,
2100
+ "fallback_rate": 0.0,
2101
+ "hidden_chart_oracle_success": 1.0,
2102
+ "hidden_chart_oracle_utility": 2.0,
2103
+ "outcome_ptr": 1.0,
2104
+ "predicted_margin": 0.783211815160911,
2105
+ "proposal_oracle_success": 1.0,
2106
+ "proposal_oracle_utility": 2.0,
2107
+ "selected_candidate_index": 15.0,
2108
+ "selected_success": 0.0,
2109
+ "selected_success_gain_over_base": 0.0,
2110
+ "selected_utility": 0.04792279005050659,
2111
+ "selected_utility_gain_over_base": -0.13379275798797607,
2112
+ "selector_regret": 1.9520772099494934,
2113
+ "success_selector_gap": 1.0,
2114
+ "success_support_gap": 0.0,
2115
+ "support_gap": 0.0,
2116
+ "tau": 0.461885278016425
2117
+ },
2118
+ "585904130/train2": {
2119
+ "base_success": 0.0,
2120
+ "base_utility": 0.18171554803848267,
2121
+ "coverage": 1.0,
2122
+ "execute_generated": 1.0,
2123
+ "fallback_rate": 0.0,
2124
+ "hidden_chart_oracle_success": 1.0,
2125
+ "hidden_chart_oracle_utility": 2.0,
2126
+ "outcome_ptr": 1.0,
2127
+ "predicted_margin": 0.783211815160911,
2128
+ "proposal_oracle_success": 1.0,
2129
+ "proposal_oracle_utility": 2.0,
2130
+ "selected_candidate_index": 1.0,
2131
+ "selected_success": 1.0,
2132
+ "selected_success_gain_over_base": 1.0,
2133
+ "selected_utility": 2.0,
2134
+ "selected_utility_gain_over_base": 1.8182844519615173,
2135
+ "selector_regret": 0.0,
2136
+ "success_selector_gap": 0.0,
2137
+ "success_support_gap": 0.0,
2138
+ "support_gap": 0.0,
2139
+ "tau": 0.461885278016425
2140
+ },
2141
+ "608016290/train0": {
2142
+ "base_success": 0.0,
2143
+ "base_utility": 0.725701630115509,
2144
+ "coverage": 0.0,
2145
+ "execute_generated": 0.0,
2146
+ "fallback_rate": 1.0,
2147
+ "hidden_chart_oracle_success": 1.0,
2148
+ "hidden_chart_oracle_utility": 2.0,
2149
+ "outcome_ptr": 1.0,
2150
+ "predicted_margin": -0.26553206302513555,
2151
+ "proposal_oracle_success": 1.0,
2152
+ "proposal_oracle_utility": 2.0,
2153
+ "selected_candidate_index": 2.0,
2154
+ "selected_success": 0.0,
2155
+ "selected_success_gain_over_base": 0.0,
2156
+ "selected_utility": 0.725701630115509,
2157
+ "selected_utility_gain_over_base": 0.0,
2158
+ "selector_regret": 1.274298369884491,
2159
+ "success_selector_gap": 1.0,
2160
+ "success_support_gap": 0.0,
2161
+ "support_gap": 0.0,
2162
+ "tau": 0.461885278016425
2163
+ },
2164
+ "608016290/train1": {
2165
+ "base_success": 0.0,
2166
+ "base_utility": 0.725701630115509,
2167
+ "coverage": 0.0,
2168
+ "execute_generated": 0.0,
2169
+ "fallback_rate": 1.0,
2170
+ "hidden_chart_oracle_success": 1.0,
2171
+ "hidden_chart_oracle_utility": 2.0,
2172
+ "outcome_ptr": 1.0,
2173
+ "predicted_margin": -0.26553206302513555,
2174
+ "proposal_oracle_success": 1.0,
2175
+ "proposal_oracle_utility": 2.0,
2176
+ "selected_candidate_index": 1.0,
2177
+ "selected_success": 0.0,
2178
+ "selected_success_gain_over_base": 0.0,
2179
+ "selected_utility": 0.725701630115509,
2180
+ "selected_utility_gain_over_base": 0.0,
2181
+ "selector_regret": 1.274298369884491,
2182
+ "success_selector_gap": 1.0,
2183
+ "success_support_gap": 0.0,
2184
+ "support_gap": 0.0,
2185
+ "tau": 0.461885278016425
2186
+ },
2187
+ "608016290/train2": {
2188
+ "base_success": 0.0,
2189
+ "base_utility": 0.725701630115509,
2190
+ "coverage": 0.0,
2191
+ "execute_generated": 0.0,
2192
+ "fallback_rate": 1.0,
2193
+ "hidden_chart_oracle_success": 1.0,
2194
+ "hidden_chart_oracle_utility": 2.0,
2195
+ "outcome_ptr": 0.0,
2196
+ "predicted_margin": -0.26553206302513555,
2197
+ "proposal_oracle_success": 0.0,
2198
+ "proposal_oracle_utility": 0.7117276191711426,
2199
+ "selected_candidate_index": 14.0,
2200
+ "selected_success": 0.0,
2201
+ "selected_success_gain_over_base": 0.0,
2202
+ "selected_utility": 0.725701630115509,
2203
+ "selected_utility_gain_over_base": 0.0,
2204
+ "selector_regret": 0.0,
2205
+ "success_selector_gap": 0.0,
2206
+ "success_support_gap": 1.0,
2207
+ "support_gap": 1.2882723808288574,
2208
+ "tau": 0.461885278016425
2209
+ },
2210
+ "626486813/train0": {
2211
+ "base_success": 0.0,
2212
+ "base_utility": 0.00509834272088483,
2213
+ "coverage": 0.0,
2214
+ "execute_generated": 0.0,
2215
+ "fallback_rate": 1.0,
2216
+ "hidden_chart_oracle_success": 0.0,
2217
+ "hidden_chart_oracle_utility": 0.16558238118886948,
2218
+ "outcome_ptr": 1.0,
2219
+ "predicted_margin": -0.05953940375790834,
2220
+ "proposal_oracle_success": 0.0,
2221
+ "proposal_oracle_utility": 0.04469547560438514,
2222
+ "selected_candidate_index": 7.0,
2223
+ "selected_success": 0.0,
2224
+ "selected_success_gain_over_base": 0.0,
2225
+ "selected_utility": 0.00509834272088483,
2226
+ "selected_utility_gain_over_base": 0.0,
2227
+ "selector_regret": 0.03959713288350031,
2228
+ "success_selector_gap": 0.0,
2229
+ "success_support_gap": 0.0,
2230
+ "support_gap": 0.12161866808310151,
2231
+ "tau": 0.461885278016425
2232
+ },
2233
+ "626486813/train1": {
2234
+ "base_success": 0.0,
2235
+ "base_utility": 0.00509834272088483,
2236
+ "coverage": 0.0,
2237
+ "execute_generated": 0.0,
2238
+ "fallback_rate": 1.0,
2239
+ "hidden_chart_oracle_success": 0.0,
2240
+ "hidden_chart_oracle_utility": 0.16558238118886948,
2241
+ "outcome_ptr": 1.0,
2242
+ "predicted_margin": 0.18621980420068318,
2243
+ "proposal_oracle_success": 0.0,
2244
+ "proposal_oracle_utility": 0.035827874671667814,
2245
+ "selected_candidate_index": 7.5,
2246
+ "selected_success": 0.0,
2247
+ "selected_success_gain_over_base": 0.0,
2248
+ "selected_utility": 0.00509834272088483,
2249
+ "selected_utility_gain_over_base": 0.0,
2250
+ "selector_regret": 0.030729531950782984,
2251
+ "success_selector_gap": 0.0,
2252
+ "success_support_gap": 0.0,
2253
+ "support_gap": 0.12975450651720166,
2254
+ "tau": 0.461885278016425
2255
+ },
2256
+ "626486813/train2": {
2257
+ "base_success": 0.0,
2258
+ "base_utility": 0.00509834272088483,
2259
+ "coverage": 0.0,
2260
+ "execute_generated": 0.0,
2261
+ "fallback_rate": 1.0,
2262
+ "hidden_chart_oracle_success": 0.0,
2263
+ "hidden_chart_oracle_utility": 0.16558238118886948,
2264
+ "outcome_ptr": 1.0,
2265
+ "predicted_margin": 0.1759229406142279,
2266
+ "proposal_oracle_success": 0.0,
2267
+ "proposal_oracle_utility": 0.04489925503730774,
2268
+ "selected_candidate_index": 3.0,
2269
+ "selected_success": 0.0,
2270
+ "selected_success_gain_over_base": 0.0,
2271
+ "selected_utility": 0.00509834272088483,
2272
+ "selected_utility_gain_over_base": 0.0,
2273
+ "selector_regret": 0.03980091231642291,
2274
+ "success_selector_gap": 0.0,
2275
+ "success_support_gap": 0.0,
2276
+ "support_gap": 0.12068312615156174,
2277
+ "tau": 0.461885278016425
2278
+ },
2279
+ "636680137/train0": {
2280
+ "base_success": 0.0,
2281
+ "base_utility": 0.08313504606485367,
2282
+ "coverage": 1.0,
2283
+ "execute_generated": 1.0,
2284
+ "fallback_rate": 0.0,
2285
+ "hidden_chart_oracle_success": 0.0,
2286
+ "hidden_chart_oracle_utility": 0.17056432366371155,
2287
+ "outcome_ptr": 1.0,
2288
+ "predicted_margin": 0.730922283654924,
2289
+ "proposal_oracle_success": 0.0,
2290
+ "proposal_oracle_utility": 0.08826130628585815,
2291
+ "selected_candidate_index": 1.0,
2292
+ "selected_success": 0.0,
2293
+ "selected_success_gain_over_base": 0.0,
2294
+ "selected_utility": 0.054690223187208176,
2295
+ "selected_utility_gain_over_base": -0.028444822877645493,
2296
+ "selector_regret": 0.03357108309864998,
2297
+ "success_selector_gap": 0.0,
2298
+ "success_support_gap": 0.0,
2299
+ "support_gap": 0.0823030173778534,
2300
+ "tau": 0.461885278016425
2301
+ },
2302
+ "636680137/train1": {
2303
+ "base_success": 0.0,
2304
+ "base_utility": 0.08313504606485367,
2305
+ "coverage": 1.0,
2306
+ "execute_generated": 1.0,
2307
+ "fallback_rate": 0.0,
2308
+ "hidden_chart_oracle_success": 0.0,
2309
+ "hidden_chart_oracle_utility": 0.17056432366371155,
2310
+ "outcome_ptr": 0.0,
2311
+ "predicted_margin": 0.5715662758164363,
2312
+ "proposal_oracle_success": 0.0,
2313
+ "proposal_oracle_utility": 0.06741118431091309,
2314
+ "selected_candidate_index": 2.0,
2315
+ "selected_success": 0.0,
2316
+ "selected_success_gain_over_base": 0.0,
2317
+ "selected_utility": 0.01698315143585205,
2318
+ "selected_utility_gain_over_base": -0.06615189462900162,
2319
+ "selector_regret": 0.050428032875061035,
2320
+ "success_selector_gap": 0.0,
2321
+ "success_support_gap": 0.0,
2322
+ "support_gap": 0.10315313935279846,
2323
+ "tau": 0.461885278016425
2324
+ },
2325
+ "636680137/train2": {
2326
+ "base_success": 0.0,
2327
+ "base_utility": 0.08313504606485367,
2328
+ "coverage": 1.0,
2329
+ "execute_generated": 1.0,
2330
+ "fallback_rate": 0.0,
2331
+ "hidden_chart_oracle_success": 0.0,
2332
+ "hidden_chart_oracle_utility": 0.17056432366371155,
2333
+ "outcome_ptr": 0.0,
2334
+ "predicted_margin": 0.730922283654924,
2335
+ "proposal_oracle_success": 0.0,
2336
+ "proposal_oracle_utility": 0.0756516233086586,
2337
+ "selected_candidate_index": 0.0,
2338
+ "selected_success": 0.0,
2339
+ "selected_success_gain_over_base": 0.0,
2340
+ "selected_utility": 0.02323226071894169,
2341
+ "selected_utility_gain_over_base": -0.05990278534591198,
2342
+ "selector_regret": 0.05241936258971691,
2343
+ "success_selector_gap": 0.0,
2344
+ "success_support_gap": 0.0,
2345
+ "support_gap": 0.09491270035505295,
2346
+ "tau": 0.461885278016425
2347
+ },
2348
+ "655884605/train0": {
2349
+ "base_success": 0.0,
2350
+ "base_utility": 0.16300667822360992,
2351
+ "coverage": 1.0,
2352
+ "execute_generated": 1.0,
2353
+ "fallback_rate": 0.0,
2354
+ "hidden_chart_oracle_success": 1.0,
2355
+ "hidden_chart_oracle_utility": 2.0,
2356
+ "outcome_ptr": 0.0,
2357
+ "predicted_margin": 0.5653215567693661,
2358
+ "proposal_oracle_success": 0.0,
2359
+ "proposal_oracle_utility": 0.13732868432998657,
2360
+ "selected_candidate_index": 5.0,
2361
+ "selected_success": 0.0,
2362
+ "selected_success_gain_over_base": 0.0,
2363
+ "selected_utility": 0.11142203211784363,
2364
+ "selected_utility_gain_over_base": -0.051584646105766296,
2365
+ "selector_regret": 0.025906652212142944,
2366
+ "success_selector_gap": 0.0,
2367
+ "success_support_gap": 1.0,
2368
+ "support_gap": 1.8626713156700134,
2369
+ "tau": 0.461885278016425
2370
+ },
2371
+ "655884605/train1": {
2372
+ "base_success": 0.0,
2373
+ "base_utility": 0.16300667822360992,
2374
+ "coverage": 1.0,
2375
+ "execute_generated": 1.0,
2376
+ "fallback_rate": 0.0,
2377
+ "hidden_chart_oracle_success": 1.0,
2378
+ "hidden_chart_oracle_utility": 2.0,
2379
+ "outcome_ptr": 1.0,
2380
+ "predicted_margin": 0.558522723271604,
2381
+ "proposal_oracle_success": 0.0,
2382
+ "proposal_oracle_utility": 0.8128758072853088,
2383
+ "selected_candidate_index": 2.0,
2384
+ "selected_success": 0.0,
2385
+ "selected_success_gain_over_base": 0.0,
2386
+ "selected_utility": 0.13658663630485535,
2387
+ "selected_utility_gain_over_base": -0.026420041918754578,
2388
+ "selector_regret": 0.6762891709804535,
2389
+ "success_selector_gap": 0.0,
2390
+ "success_support_gap": 1.0,
2391
+ "support_gap": 1.1871241927146912,
2392
+ "tau": 0.461885278016425
2393
+ },
2394
+ "655884605/train2": {
2395
+ "base_success": 0.0,
2396
+ "base_utility": 0.16300667822360992,
2397
+ "coverage": 1.0,
2398
+ "execute_generated": 1.0,
2399
+ "fallback_rate": 0.0,
2400
+ "hidden_chart_oracle_success": 1.0,
2401
+ "hidden_chart_oracle_utility": 2.0,
2402
+ "outcome_ptr": 1.0,
2403
+ "predicted_margin": 0.558522723271604,
2404
+ "proposal_oracle_success": 0.0,
2405
+ "proposal_oracle_utility": 0.21447330713272095,
2406
+ "selected_candidate_index": 3.0,
2407
+ "selected_success": 0.0,
2408
+ "selected_success_gain_over_base": 0.0,
2409
+ "selected_utility": 0.19850653409957886,
2410
+ "selected_utility_gain_over_base": 0.03549985587596893,
2411
+ "selector_regret": 0.01596677303314209,
2412
+ "success_selector_gap": 0.0,
2413
+ "success_support_gap": 1.0,
2414
+ "support_gap": 1.785526692867279,
2415
+ "tau": 0.461885278016425
2416
+ },
2417
+ "672328262/train0": {
2418
+ "base_success": 1.0,
2419
+ "base_utility": 2.0,
2420
+ "coverage": 0.0,
2421
+ "execute_generated": 0.0,
2422
+ "fallback_rate": 1.0,
2423
+ "hidden_chart_oracle_success": 1.0,
2424
+ "hidden_chart_oracle_utility": 2.0,
2425
+ "outcome_ptr": 0.0,
2426
+ "predicted_margin": 0.37033219487548785,
2427
+ "proposal_oracle_success": 1.0,
2428
+ "proposal_oracle_utility": 2.0,
2429
+ "selected_candidate_index": 1.0,
2430
+ "selected_success": 1.0,
2431
+ "selected_success_gain_over_base": 0.0,
2432
+ "selected_utility": 2.0,
2433
+ "selected_utility_gain_over_base": 0.0,
2434
+ "selector_regret": 0.0,
2435
+ "success_selector_gap": 0.0,
2436
+ "success_support_gap": 0.0,
2437
+ "support_gap": 0.0,
2438
+ "tau": 0.461885278016425
2439
+ },
2440
+ "672328262/train1": {
2441
+ "base_success": 1.0,
2442
+ "base_utility": 2.0,
2443
+ "coverage": 0.0,
2444
+ "execute_generated": 0.0,
2445
+ "fallback_rate": 1.0,
2446
+ "hidden_chart_oracle_success": 1.0,
2447
+ "hidden_chart_oracle_utility": 2.0,
2448
+ "outcome_ptr": 0.0,
2449
+ "predicted_margin": -0.09793626053678911,
2450
+ "proposal_oracle_success": 1.0,
2451
+ "proposal_oracle_utility": 2.0,
2452
+ "selected_candidate_index": 1.0,
2453
+ "selected_success": 1.0,
2454
+ "selected_success_gain_over_base": 0.0,
2455
+ "selected_utility": 2.0,
2456
+ "selected_utility_gain_over_base": 0.0,
2457
+ "selector_regret": 0.0,
2458
+ "success_selector_gap": 0.0,
2459
+ "success_support_gap": 0.0,
2460
+ "support_gap": 0.0,
2461
+ "tau": 0.461885278016425
2462
+ },
2463
+ "672328262/train2": {
2464
+ "base_success": 1.0,
2465
+ "base_utility": 2.0,
2466
+ "coverage": 0.0,
2467
+ "execute_generated": 0.0,
2468
+ "fallback_rate": 1.0,
2469
+ "hidden_chart_oracle_success": 1.0,
2470
+ "hidden_chart_oracle_utility": 2.0,
2471
+ "outcome_ptr": 0.0,
2472
+ "predicted_margin": 0.37033219487548785,
2473
+ "proposal_oracle_success": 1.0,
2474
+ "proposal_oracle_utility": 2.0,
2475
+ "selected_candidate_index": 2.0,
2476
+ "selected_success": 1.0,
2477
+ "selected_success_gain_over_base": 0.0,
2478
+ "selected_utility": 2.0,
2479
+ "selected_utility_gain_over_base": 0.0,
2480
+ "selector_regret": 0.0,
2481
+ "success_selector_gap": 0.0,
2482
+ "success_support_gap": 0.0,
2483
+ "support_gap": 0.0,
2484
+ "tau": 0.461885278016425
2485
+ },
2486
+ "701667486/train0": {
2487
+ "base_success": 1.0,
2488
+ "base_utility": 2.0,
2489
+ "coverage": 1.0,
2490
+ "execute_generated": 1.0,
2491
+ "fallback_rate": 0.0,
2492
+ "hidden_chart_oracle_success": 1.0,
2493
+ "hidden_chart_oracle_utility": 2.0,
2494
+ "outcome_ptr": 0.0,
2495
+ "predicted_margin": 1.5199957201646186,
2496
+ "proposal_oracle_success": 1.0,
2497
+ "proposal_oracle_utility": 2.0,
2498
+ "selected_candidate_index": 1.0,
2499
+ "selected_success": 1.0,
2500
+ "selected_success_gain_over_base": 0.0,
2501
+ "selected_utility": 2.0,
2502
+ "selected_utility_gain_over_base": 0.0,
2503
+ "selector_regret": 0.0,
2504
+ "success_selector_gap": 0.0,
2505
+ "success_support_gap": 0.0,
2506
+ "support_gap": 0.0,
2507
+ "tau": 0.461885278016425
2508
+ },
2509
+ "701667486/train1": {
2510
+ "base_success": 1.0,
2511
+ "base_utility": 2.0,
2512
+ "coverage": 1.0,
2513
+ "execute_generated": 1.0,
2514
+ "fallback_rate": 0.0,
2515
+ "hidden_chart_oracle_success": 1.0,
2516
+ "hidden_chart_oracle_utility": 2.0,
2517
+ "outcome_ptr": 0.0,
2518
+ "predicted_margin": 1.5199957201646186,
2519
+ "proposal_oracle_success": 1.0,
2520
+ "proposal_oracle_utility": 2.0,
2521
+ "selected_candidate_index": 6.0,
2522
+ "selected_success": 1.0,
2523
+ "selected_success_gain_over_base": 0.0,
2524
+ "selected_utility": 2.0,
2525
+ "selected_utility_gain_over_base": 0.0,
2526
+ "selector_regret": 0.0,
2527
+ "success_selector_gap": 0.0,
2528
+ "success_support_gap": 0.0,
2529
+ "support_gap": 0.0,
2530
+ "tau": 0.461885278016425
2531
+ },
2532
+ "701667486/train2": {
2533
+ "base_success": 1.0,
2534
+ "base_utility": 2.0,
2535
+ "coverage": 1.0,
2536
+ "execute_generated": 1.0,
2537
+ "fallback_rate": 0.0,
2538
+ "hidden_chart_oracle_success": 1.0,
2539
+ "hidden_chart_oracle_utility": 2.0,
2540
+ "outcome_ptr": 0.0,
2541
+ "predicted_margin": 1.5199957201646186,
2542
+ "proposal_oracle_success": 1.0,
2543
+ "proposal_oracle_utility": 2.0,
2544
+ "selected_candidate_index": 1.0,
2545
+ "selected_success": 1.0,
2546
+ "selected_success_gain_over_base": 0.0,
2547
+ "selected_utility": 2.0,
2548
+ "selected_utility_gain_over_base": 0.0,
2549
+ "selector_regret": 0.0,
2550
+ "success_selector_gap": 0.0,
2551
+ "success_support_gap": 0.0,
2552
+ "support_gap": 0.0,
2553
+ "tau": 0.461885278016425
2554
+ },
2555
+ "747260437/train0": {
2556
+ "base_success": 1.0,
2557
+ "base_utility": 2.0,
2558
+ "coverage": 0.0,
2559
+ "execute_generated": 0.0,
2560
+ "fallback_rate": 1.0,
2561
+ "hidden_chart_oracle_success": 0.0,
2562
+ "hidden_chart_oracle_utility": 0.2786632478237152,
2563
+ "outcome_ptr": 0.0,
2564
+ "predicted_margin": 0.11147982914005515,
2565
+ "proposal_oracle_success": 0.0,
2566
+ "proposal_oracle_utility": 0.10839563608169556,
2567
+ "selected_candidate_index": 15.0,
2568
+ "selected_success": 1.0,
2569
+ "selected_success_gain_over_base": 0.0,
2570
+ "selected_utility": 2.0,
2571
+ "selected_utility_gain_over_base": 0.0,
2572
+ "selector_regret": 0.0,
2573
+ "success_selector_gap": 0.0,
2574
+ "success_support_gap": 0.0,
2575
+ "support_gap": 0.17026761174201965,
2576
+ "tau": 0.461885278016425
2577
+ },
2578
+ "747260437/train1": {
2579
+ "base_success": 1.0,
2580
+ "base_utility": 2.0,
2581
+ "coverage": 0.0,
2582
+ "execute_generated": 0.0,
2583
+ "fallback_rate": 1.0,
2584
+ "hidden_chart_oracle_success": 0.0,
2585
+ "hidden_chart_oracle_utility": 0.2786632478237152,
2586
+ "outcome_ptr": 0.0,
2587
+ "predicted_margin": 0.15923551786737147,
2588
+ "proposal_oracle_success": 0.0,
2589
+ "proposal_oracle_utility": 0.06328952312469482,
2590
+ "selected_candidate_index": 0.0,
2591
+ "selected_success": 1.0,
2592
+ "selected_success_gain_over_base": 0.0,
2593
+ "selected_utility": 2.0,
2594
+ "selected_utility_gain_over_base": 0.0,
2595
+ "selector_regret": 0.0,
2596
+ "success_selector_gap": 0.0,
2597
+ "success_support_gap": 0.0,
2598
+ "support_gap": 0.21537372469902039,
2599
+ "tau": 0.461885278016425
2600
+ },
2601
+ "747260437/train2": {
2602
+ "base_success": 1.0,
2603
+ "base_utility": 2.0,
2604
+ "coverage": 0.0,
2605
+ "execute_generated": 0.0,
2606
+ "fallback_rate": 1.0,
2607
+ "hidden_chart_oracle_success": 0.0,
2608
+ "hidden_chart_oracle_utility": 0.2786632478237152,
2609
+ "outcome_ptr": 0.0,
2610
+ "predicted_margin": 0.15923551786737147,
2611
+ "proposal_oracle_success": 0.0,
2612
+ "proposal_oracle_utility": 0.12510187923908234,
2613
+ "selected_candidate_index": 4.0,
2614
+ "selected_success": 1.0,
2615
+ "selected_success_gain_over_base": 0.0,
2616
+ "selected_utility": 2.0,
2617
+ "selected_utility_gain_over_base": 0.0,
2618
+ "selector_regret": 0.0,
2619
+ "success_selector_gap": 0.0,
2620
+ "success_support_gap": 0.0,
2621
+ "support_gap": 0.15356136858463287,
2622
+ "tau": 0.461885278016425
2623
+ },
2624
+ "750448296/train0": {
2625
+ "base_success": 0.0,
2626
+ "base_utility": 0.11759314934412639,
2627
+ "coverage": 1.0,
2628
+ "execute_generated": 1.0,
2629
+ "fallback_rate": 0.0,
2630
+ "hidden_chart_oracle_success": 0.6666666666666666,
2631
+ "hidden_chart_oracle_utility": 1.4079964955647786,
2632
+ "outcome_ptr": 0.6666666666666666,
2633
+ "predicted_margin": 0.8358427629109161,
2634
+ "proposal_oracle_success": 0.3333333333333333,
2635
+ "proposal_oracle_utility": 0.7526603279014429,
2636
+ "selected_candidate_index": 2.6666666666666665,
2637
+ "selected_success": 0.3333333333333333,
2638
+ "selected_success_gain_over_base": 0.3333333333333333,
2639
+ "selected_utility": 0.7116076741367579,
2640
+ "selected_utility_gain_over_base": 0.5940145247926315,
2641
+ "selector_regret": 0.041052653764685,
2642
+ "success_selector_gap": 0.0,
2643
+ "success_support_gap": 0.3333333333333333,
2644
+ "support_gap": 0.6553361676633358,
2645
+ "tau": 0.4618852780164249
2646
+ },
2647
+ "750448296/train1": {
2648
+ "base_success": 0.0,
2649
+ "base_utility": 0.11759314934412639,
2650
+ "coverage": 0.6666666666666666,
2651
+ "execute_generated": 0.6666666666666666,
2652
+ "fallback_rate": 0.3333333333333333,
2653
+ "hidden_chart_oracle_success": 0.6666666666666666,
2654
+ "hidden_chart_oracle_utility": 1.4079964955647786,
2655
+ "outcome_ptr": 1.0,
2656
+ "predicted_margin": 0.6027052046516154,
2657
+ "proposal_oracle_success": 0.6666666666666666,
2658
+ "proposal_oracle_utility": 1.3705032741030057,
2659
+ "selected_candidate_index": 1.6666666666666667,
2660
+ "selected_success": 0.3333333333333333,
2661
+ "selected_success_gain_over_base": 0.3333333333333333,
2662
+ "selected_utility": 0.7338168621063232,
2663
+ "selected_utility_gain_over_base": 0.6162237127621969,
2664
+ "selector_regret": 0.6366864119966825,
2665
+ "success_selector_gap": 0.3333333333333333,
2666
+ "success_support_gap": 0.0,
2667
+ "support_gap": 0.03749322146177292,
2668
+ "tau": 0.4618852780164249
2669
+ },
2670
+ "750448296/train2": {
2671
+ "base_success": 0.0,
2672
+ "base_utility": 0.11759314934412639,
2673
+ "coverage": 0.6666666666666666,
2674
+ "execute_generated": 0.6666666666666666,
2675
+ "fallback_rate": 0.3333333333333333,
2676
+ "hidden_chart_oracle_success": 0.6666666666666666,
2677
+ "hidden_chart_oracle_utility": 1.4079964955647786,
2678
+ "outcome_ptr": 0.6666666666666666,
2679
+ "predicted_margin": 0.5885351499944433,
2680
+ "proposal_oracle_success": 0.6666666666666666,
2681
+ "proposal_oracle_utility": 1.3504486680030823,
2682
+ "selected_candidate_index": 0.6666666666666666,
2683
+ "selected_success": 0.6666666666666666,
2684
+ "selected_success_gain_over_base": 0.6666666666666666,
2685
+ "selected_utility": 1.3643324971199036,
2686
+ "selected_utility_gain_over_base": 1.246739347775777,
2687
+ "selector_regret": 0.0,
2688
+ "success_selector_gap": 0.0,
2689
+ "success_support_gap": 0.0,
2690
+ "support_gap": 0.05754782756169637,
2691
+ "tau": 0.4618852780164249
2692
+ },
2693
+ "767747424/train0": {
2694
+ "base_success": 0.25,
2695
+ "base_utility": 0.5731993420049548,
2696
+ "coverage": 0.5,
2697
+ "execute_generated": 0.5,
2698
+ "fallback_rate": 0.5,
2699
+ "hidden_chart_oracle_success": 0.25,
2700
+ "hidden_chart_oracle_utility": 0.6737305596470833,
2701
+ "outcome_ptr": 0.5,
2702
+ "predicted_margin": 0.5165554186299326,
2703
+ "proposal_oracle_success": 0.5,
2704
+ "proposal_oracle_utility": 1.0424291878007352,
2705
+ "selected_candidate_index": 11.25,
2706
+ "selected_success": 0.0,
2707
+ "selected_success_gain_over_base": -0.25,
2708
+ "selected_utility": 0.039441852990421467,
2709
+ "selected_utility_gain_over_base": -0.5337574890145333,
2710
+ "selector_regret": 1.0117176117637428,
2711
+ "success_selector_gap": 0.5,
2712
+ "success_support_gap": 0.0,
2713
+ "support_gap": 0.06562284985557199,
2714
+ "tau": 0.461885278016425
2715
+ },
2716
+ "767747424/train1": {
2717
+ "base_success": 0.25,
2718
+ "base_utility": 0.5731993420049548,
2719
+ "coverage": 0.25,
2720
+ "execute_generated": 0.25,
2721
+ "fallback_rate": 0.75,
2722
+ "hidden_chart_oracle_success": 0.25,
2723
+ "hidden_chart_oracle_utility": 0.6737305596470833,
2724
+ "outcome_ptr": 0.5,
2725
+ "predicted_margin": 0.51883899857906,
2726
+ "proposal_oracle_success": 0.5,
2727
+ "proposal_oracle_utility": 1.0629530912265182,
2728
+ "selected_candidate_index": 8.75,
2729
+ "selected_success": 0.25,
2730
+ "selected_success_gain_over_base": 0.0,
2731
+ "selected_utility": 0.5731993420049548,
2732
+ "selected_utility_gain_over_base": 0.0,
2733
+ "selector_regret": 0.4925890024751425,
2734
+ "success_selector_gap": 0.25,
2735
+ "success_support_gap": 0.0,
2736
+ "support_gap": 0.045098946429789066,
2737
+ "tau": 0.461885278016425
2738
+ },
2739
+ "767747424/train2": {
2740
+ "base_success": 0.25,
2741
+ "base_utility": 0.5731993420049548,
2742
+ "coverage": 0.5,
2743
+ "execute_generated": 0.5,
2744
+ "fallback_rate": 0.5,
2745
+ "hidden_chart_oracle_success": 0.25,
2746
+ "hidden_chart_oracle_utility": 0.6737305596470833,
2747
+ "outcome_ptr": 0.5,
2748
+ "predicted_margin": 0.45152492482904577,
2749
+ "proposal_oracle_success": 0.5,
2750
+ "proposal_oracle_utility": 1.0671670949086547,
2751
+ "selected_candidate_index": 3.25,
2752
+ "selected_success": 0.0,
2753
+ "selected_success_gain_over_base": -0.25,
2754
+ "selected_utility": 0.08047536047524773,
2755
+ "selected_utility_gain_over_base": -0.4927239815297071,
2756
+ "selector_regret": 0.9877707070263568,
2757
+ "success_selector_gap": 0.5,
2758
+ "success_support_gap": 0.0,
2759
+ "support_gap": 0.04088494274765253,
2760
+ "tau": 0.461885278016425
2761
+ }
2762
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/metrics_by_task.json ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "LiftPegUpright-v1": {
3
+ "base_success": 0.2,
4
+ "base_utility": 0.9498705387115478,
5
+ "coverage": 0.0,
6
+ "execute_generated": 0.0,
7
+ "fallback_rate": 1.0,
8
+ "hidden_chart_oracle_success": 1.0,
9
+ "hidden_chart_oracle_utility": 2.0,
10
+ "outcome_ptr": 0.5333333333333333,
11
+ "predicted_margin": -0.052165254117878486,
12
+ "proposal_oracle_success": 0.6,
13
+ "proposal_oracle_utility": 1.4122399787108104,
14
+ "selected_candidate_index": 3.466666666666667,
15
+ "selected_success": 0.2,
16
+ "selected_success_gain_over_base": 0.0,
17
+ "selected_utility": 0.9498705387115478,
18
+ "selected_utility_gain_over_base": 0.0,
19
+ "selector_regret": 0.5146939317385356,
20
+ "success_selector_gap": 0.4,
21
+ "success_support_gap": 0.4,
22
+ "support_gap": 0.5877600212891897,
23
+ "tau": 0.46188527801642487
24
+ },
25
+ "PickCube-v1": {
26
+ "base_success": 0.0,
27
+ "base_utility": 0.5839591026306152,
28
+ "coverage": 0.0,
29
+ "execute_generated": 0.0,
30
+ "fallback_rate": 1.0,
31
+ "hidden_chart_oracle_success": 1.0,
32
+ "hidden_chart_oracle_utility": 2.0,
33
+ "outcome_ptr": 0.0,
34
+ "predicted_margin": -0.07343996006022091,
35
+ "proposal_oracle_success": 0.0,
36
+ "proposal_oracle_utility": 0.13425676027933756,
37
+ "selected_candidate_index": 6.666666666666667,
38
+ "selected_success": 0.0,
39
+ "selected_success_gain_over_base": 0.0,
40
+ "selected_utility": 0.5839591026306152,
41
+ "selected_utility_gain_over_base": 0.0,
42
+ "selector_regret": 0.0,
43
+ "success_selector_gap": 0.0,
44
+ "success_support_gap": 1.0,
45
+ "support_gap": 1.8657432397206624,
46
+ "tau": 0.4618852780164249
47
+ },
48
+ "PullCube-v1": {
49
+ "base_success": 0.36,
50
+ "base_utility": 0.7751704612234608,
51
+ "coverage": 0.3333333333333333,
52
+ "execute_generated": 0.3333333333333333,
53
+ "fallback_rate": 0.6666666666666666,
54
+ "hidden_chart_oracle_success": 0.52,
55
+ "hidden_chart_oracle_utility": 1.1467878860235214,
56
+ "outcome_ptr": 0.4666666666666667,
57
+ "predicted_margin": 0.3834765656041302,
58
+ "proposal_oracle_success": 0.5333333333333333,
59
+ "proposal_oracle_utility": 1.1093241602430741,
60
+ "selected_candidate_index": 4.16,
61
+ "selected_success": 0.3466666666666667,
62
+ "selected_success_gain_over_base": -0.013333333333333334,
63
+ "selected_utility": 0.7355007326719352,
64
+ "selected_utility_gain_over_base": -0.039669728551525625,
65
+ "selector_regret": 0.4549120608600788,
66
+ "success_selector_gap": 0.22666666666666666,
67
+ "success_support_gap": 0.02666666666666667,
68
+ "support_gap": 0.1075624164690574,
69
+ "tau": 0.46188527801642476
70
+ },
71
+ "PushCube-v1": {
72
+ "base_success": 0.16666666666666666,
73
+ "base_utility": 0.3924041787783305,
74
+ "coverage": 0.6666666666666666,
75
+ "execute_generated": 0.6666666666666666,
76
+ "fallback_rate": 0.3333333333333333,
77
+ "hidden_chart_oracle_success": 0.8333333333333334,
78
+ "hidden_chart_oracle_utility": 1.7015800600250561,
79
+ "outcome_ptr": 0.8333333333333334,
80
+ "predicted_margin": 0.8374506997361121,
81
+ "proposal_oracle_success": 0.8888888888888888,
82
+ "proposal_oracle_utility": 1.7857814149724112,
83
+ "selected_candidate_index": 4.555555555555555,
84
+ "selected_success": 0.4444444444444444,
85
+ "selected_success_gain_over_base": 0.2777777777777778,
86
+ "selected_utility": 0.9414158223403825,
87
+ "selected_utility_gain_over_base": 0.549011643562052,
88
+ "selector_regret": 0.8443655926320288,
89
+ "success_selector_gap": 0.4444444444444444,
90
+ "success_support_gap": 0.1111111111111111,
91
+ "support_gap": 0.21421858502758873,
92
+ "tau": 0.46188527801642487
93
+ },
94
+ "StackCube-v1": {
95
+ "base_success": 0.2727272727272727,
96
+ "base_utility": 0.7763947045261209,
97
+ "coverage": 0.36363636363636365,
98
+ "execute_generated": 0.36363636363636365,
99
+ "fallback_rate": 0.6363636363636364,
100
+ "hidden_chart_oracle_success": 1.0,
101
+ "hidden_chart_oracle_utility": 2.0,
102
+ "outcome_ptr": 0.6363636363636364,
103
+ "predicted_margin": 0.3693245330816113,
104
+ "proposal_oracle_success": 0.5151515151515151,
105
+ "proposal_oracle_utility": 1.1567951707225856,
106
+ "selected_candidate_index": 3.878787878787879,
107
+ "selected_success": 0.30303030303030304,
108
+ "selected_success_gain_over_base": 0.030303030303030304,
109
+ "selected_utility": 0.7855989084099279,
110
+ "selected_utility_gain_over_base": 0.009204203883806864,
111
+ "selector_regret": 0.3740111264315518,
112
+ "success_selector_gap": 0.21212121212121213,
113
+ "success_support_gap": 0.48484848484848486,
114
+ "support_gap": 0.8432048292774142,
115
+ "tau": 0.46188527801642476
116
+ }
117
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/report.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Nonlinear Train-Calibrated CTT Selector
2
+
3
+ Calibration rows: `432`
4
+ Fit rows: `277`
5
+ Selection rows: `155`
6
+ Eval rows: `144`
7
+ Selected model: `hgb_regressor`
8
+ Selected params: `{"l2_regularization": 0.01, "learning_rate": 0.05, "max_iter": 40, "max_leaf_nodes": 15, "random_state": 0}`
9
+ Tau: `0.461885`
10
+ Feature set: `chart_compat`
11
+ Target: `success_weighted_margin`
12
+
13
+ The model is fit on calibration-fit rows, and model/tau are selected on held-out calibration-selection rows only. Eval outcomes are used only for reporting.
14
+
15
+ | Split | Coverage | Fallback | Base success | Selected success | Proposal oracle | OutcomePTR | Success support gap | Success selector gap |
16
+ | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
17
+ | fit | 0.3357 | 0.6643 | 0.2780 | 0.4621 | 0.5343 | 0.5343 | 0.1733 | 0.1264 |
18
+ | selection | 0.3355 | 0.6645 | 0.2774 | 0.3742 | 0.5419 | 0.5677 | 0.1806 | 0.2065 |
19
+ | eval | 0.3403 | 0.6597 | 0.2917 | 0.3264 | 0.5694 | 0.5486 | 0.2014 | 0.2639 |
20
+
21
+ This is a train-calibrated selector diagnostic over already measured candidates, not a new rollout.
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/split_hash.txt ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "calibration_target_index": {
3
+ "content_hash": "67f7b4a692f7c7e71da24378dc71e3399c5d35afc14c494869eba6087b765c42",
4
+ "retrieval_index_allowed": true,
5
+ "split": "train",
6
+ "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372"
7
+ },
8
+ "eval_target_index": {
9
+ "content_hash": "0c38d73dc9bc4ab756e12006671b5d12f1a23358b281e54b3f6a9ae3cbf6e9af",
10
+ "retrieval_index_allowed": false,
11
+ "split": "test",
12
+ "split_hash": "84151df9e88c219b76751677558cc5bbbc19082bf358cea7330158e9e403e25f"
13
+ },
14
+ "source_index": {
15
+ "content_hash": "67f7b4a692f7c7e71da24378dc71e3399c5d35afc14c494869eba6087b765c42",
16
+ "retrieval_index_allowed": true,
17
+ "split": "train",
18
+ "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372"
19
+ }
20
+ }
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/table.tex ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ % Auto-generated by scripts/eval_nonlinear_dominance_selector.py
2
+ \begin{tabular}{lrrrrrrrr}
3
+ \toprule
4
+ Rows & Coverage & Fallback & Base succ. & Selected succ. & Oracle succ. & OutcomePTR & Succ. support gap & Succ. selector gap \\
5
+ \midrule
6
+ 144 & 0.3403 & 0.6597 & 0.2917 & 0.3264 & 0.5694 & 0.5486 & 0.2014 & 0.2639 \\
7
+ \bottomrule
8
+ \end{tabular}
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_success_weighted_reg_envclip_k16_train_to_test/train.log ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ trained nonlinear dominance selector on calibration-fit rows only
2
+ selected_model_type=hgb_regressor
3
+ selected_model_params={"l2_regularization": 0.01, "learning_rate": 0.05, "max_iter": 40, "max_leaf_nodes": 15, "random_state": 0}
4
+ tau=0.461885
5
+ calibration_input=runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_utility_envclip_k16_train_to_test/command.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ python scripts/eval_nonlinear_dominance_selector.py --calibration-input runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json --calibration-target-index data/cil_charts_rgb_refs/train/index.json --eval-input runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json --eval-target-index data/cil_charts_rgb_refs/test/index.json --source-index data/cil_charts_rgb_refs/train/index.json --checkpoint-template runs/ctt_residual_base_context_obs_seed{seed}/model.pt --out-dir runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_utility_envclip_k16_train_to_test --feature-set chart_compat --selector-chart-feature-mode base_context_obs --target utility_margin --k 16 --bootstrap-samples 1000
workspace/runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_utility_envclip_k16_train_to_test/config.yaml ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ bootstrap_samples: 1000
2
+ calibration_input: runs/ctt_base_context_obs_train_cal_envclip_k16_rollout_comparison/combined_measured_candidates.json
3
+ calibration_target_index: data/cil_charts_rgb_refs/train/index.json
4
+ checkpoint_template: runs/ctt_residual_base_context_obs_seed{seed}/model.pt
5
+ eval_input: runs/ctt_base_context_obs_test_envclip_k16_rollout_comparison/combined_measured_candidates.json
6
+ eval_target_index: data/cil_charts_rgb_refs/test/index.json
7
+ feature_set: chart_compat
8
+ k: 16
9
+ model_types: hgb_classifier,hgb_regressor,rf_regressor
10
+ out_dir: runs/ctt_base_context_obs_nonlinear_dominance_chartcompat_obs_utility_envclip_k16_train_to_test
11
+ seed: 0
12
+ selection_frac: 0.35
13
+ selector_chart_feature_mode: base_context_obs
14
+ source_index: data/cil_charts_rgb_refs/train/index.json
15
+ target: utility_margin