Constrained Reinforcement Learning on MuJoCo Hopper eval environment
1,610.7RewardTraCeS
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| TraCeSKnowledge Level=Zero Knowledge2025.04 | 1,610.7 | 17.1 | 2,993 | |
| TRPO-SauteKnowledge Level=Oracle2025.04 | 1,357.8 | 16.3 | — | |
| FOCOPSKnowledge Level=Oracle2025.04 | 1,330.1 | 31.9 | — | |
| CUPKnowledge Level=Oracle2025.04 | 1,273.5 | 13 | — |