Best Arm Identification on Unstructured instances K=40
5.43Samplesk-Learner
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| k-LearnerElimination Strategy=Elim. stopping + sampling2022.05 | 5.43 | 0.52 | |
| FWSElimination Strategy=Elim. stopping + sampling2022.05 | 5.47 | 1.4 | |
| LUCBElimination Strategy=Elim. stopping + sampling2022.05 | 5.49 | 0.1 | |
| LUCBElimination Strategy=No elim. (LLR)2022.05 | 5.5 | 0.11 | |
| LUCBElimination Strategy=Elim. stopping2022.05 | 5.5 | 0.1 | |
| FWSElimination Strategy=Elim. stopping2022.05 | 5.53 | 1.47 | |
| FWSElimination Strategy=No elim. (LLR)2022.05 | 5.54 | 1.48 | |
| k-LearnerElimination Strategy=Elim. stopping2022.05 | 8.38 | 0.55 | |
| k-LearnerElimination Strategy=No elim. (LLR)2022.05 | 8.55 | 0.61 | |
| Lazy TaSElimination Strategy=Elim. stopping + sampling2022.05 | 8.74 | 0.79 | |
| OracleElimination Strategy=Elim. stopping2022.05 | 11.41 | 0.04 | |
| OracleElimination Strategy=No elim. (LLR)2022.05 | 11.55 | 0.07 | |
| Lazy TaSElimination Strategy=Elim. stopping2022.05 | 12.27 | 0.7 | |
| Lazy TaSElimination Strategy=No elim. (LLR)2022.05 | 12.83 | 0.72 | |
| LUCBElimination Strategy=Elim. stopping + sampling2022.05 | 13.57 | 0.06 | |
| FWSElimination Strategy=Elim. stopping + sampling2022.05 | 13.94 | 1.16 | |
| LUCBElimination Strategy=Elim. stopping2022.05 | 14.18 | 0.06 | |
| LUCBElimination Strategy=No elim. (LLR)2022.05 | 14.2 | 0.11 | |
| FWSElimination Strategy=Elim. stopping2022.05 | 14.44 | 1.24 | |
| FWSElimination Strategy=No elim. (LLR)2022.05 | 14.5 | 1.25 | |
| k-LearnerElimination Strategy=Elim. stopping + sampling2022.05 | 14.82 | 0.4 | |
| UGapEElimination Strategy=No elim. (LLR)2022.05 | 15.13 | 0.43 | |
| UGapEElimination Strategy=Elim. stopping2022.05 | 15.13 | 0.39 | |
| k-LearnerElimination Strategy=Elim. stopping2022.05 | 18.12 | 0.44 | |
| k-LearnerElimination Strategy=No elim. (LLR)2022.05 | 18.76 | 0.49 | |
| Lazy TaSElimination Strategy=Elim. stopping + sampling2022.05 | 20.66 | 0.33 | |
| Lazy TaSElimination Strategy=Elim. stopping2022.05 | 24.78 | 0.31 | |
| Lazy TaSElimination Strategy=No elim. (LLR)2022.05 | 26.18 | 0.32 | |
| OracleElimination Strategy=Elim. stopping2022.05 | 27 | 0.04 | |
| OracleElimination Strategy=No elim. (LLR)2022.05 | 27.49 | 0.07 | |
| RacingElimination Strategy=Elim. stopping + sampling2022.05 | 34.55 | 0.01 |