Policy Optimization in CMAB on 1084 3
0.0018Mean Diff from Supervised (PV-loss)OPO-CMAB
Evaluation Results
| Method | Links | |
|---|---|---|
| OPO-CMABEvaluation Protocol=Averaged-across-permutations2026.02 | 0.0018 | |
| RegCBEvaluation Protocol=Averaged-across-permutations2026.02 | 0.0095 | |
| FastCBEvaluation Protocol=Averaged-across-permutations2026.02 | 0.0828 | |
| SquareCBEvaluation Protocol=Averaged-across-permutations2026.02 | 0.145 | |
| AdaCBEvaluation Protocol=Averaged-across-permutations2026.02 | 0.215 |