Constrained Reinforcement Learning on MuJoCo Walker2d (eval)
2,455.3RewardFOCOPS
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| FOCOPSKnowledge Level=Oracle2025.04 | 2,455.3 | 8.3 | — | |
| CUPKnowledge Level=Oracle2025.04 | 2,388.1 | 6.6 | — | |
| TraCeSKnowledge Level=Zero Knowledge2025.04 | 2,211.2 | 24.1 | 5,735 | |
| TRPO-SauteKnowledge Level=Oracle2025.04 | 2,138.7 | 18.1 | — |