Offline Reinforcement Learning on Walker2D Medium-Expert JointNoise Shift
118.564Average ReturnREAG*MV
Evaluation Results
| Method | Links | |
|---|---|---|
| REAG*MVFramework=QT2024.10 | 118.564 | |
| REAG*MVFramework=Reinformer2024.10 | 118.218 | |
| REAG*DaraFramework=Reinformer2024.10 | 117.765 | |
| 1T10SFramework=Reinformer2024.10 | 117.36 | |
| 1T10SFramework=QT2024.10 | 116.149 | |
| REAG*DaraFramework=QT2024.10 | 116.115 | |
| 1T10SFramework=DT2024.10 | 115.746 | |
| REAG*DaraFramework=DT2024.10 | 111.236 | |
| REAG*MVFramework=DT2024.10 | 111.06 |