Offline Reinforcement Learning on D4RL MuJoCo Walker2d-m medium v2
86.8Average Normalized ScoreDOGE
Evaluation results
| Method | Links | |
|---|---|---|
| DOGERatio=1, Size=1M2023.06 | 86.8 | |
| TD3+BCRatio=1, Size=1M2023.06 | 83.7 | |
| IQLRatio=1, Size=1M2023.06 | 78.3 | |
| TSRLRatio=1, Size=1M2023.06 | 77.5 | |
| BCRatio=1, Size=1M2023.06 | 75.3 | |
| CQLRatio=1, Size=1M2023.06 | 72.5 | |
| TSRLRatio=1/100, Size=10k2023.06 | 49.7 | |
| DOGERatio=1/100, Size=10k2023.06 | 45.1 | |
| IQLRatio=1/100, Size=10k2023.06 | 22.5 | |
| CQLRatio=1/100, Size=10k2023.06 | 18.8 | |
| MOPORatio=1, Size=1M2023.06 | 17.8 | |
| BCRatio=1/100, Size=10k2023.06 | 15.8 | |
| TD3+BCRatio=1/100, Size=10k2023.06 | 7.4 | |
| MOPORatio=1/100, Size=10k2023.06 | 3.1 |