Offline Reinforcement Learning on Hopper Medium-Expert v2
77.21Average ScoreRLDP
Evaluation results
| Method | Links | |
|---|---|---|
| RLDPSetting=low-coverage2026.03 | 77.21 | |
| HILPSetting=low-coverage2026.03 | 68.18 | |
| TRACERcorruption_type=random simultaneous corruptions2024.11 | 55.5 | |
| FBSetting=low-coverage2026.03 | 54.64 | |
| PSMSetting=low-coverage2026.03 | 14.59 | |
| RIQLcorruption_type=random simultaneous corruptions2024.11 | 3.9 | |
| IQLcorruption_type=random simultaneous corruptions2024.11 | 1.4 |