Policy Optimization in CMAB on 905 2
0.0333Mean Diff from Supervised (PV-loss)RegCB
Evaluation Results
| Method | Links | |
|---|---|---|
| RegCBEvaluation Protocol=Averaged-across-permutations2026.02 | 0.0333 | |
| AdaCBEvaluation Protocol=Averaged-across-permutations2026.02 | 0.0641 | |
| FastCBEvaluation Protocol=Averaged-across-permutations2026.02 | 0.0667 | |
| SquareCBEvaluation Protocol=Averaged-across-permutations2026.02 | 0.0667 | |
| OPO-CMABEvaluation Protocol=Averaged-across-permutations2026.02 | 0.1641 |