Offline Reinforcement Learning on Walker2D Medium-Replay BodyMass Shift
87.491Average ReturnREAG*MV
Evaluation Results
| Method | Links | |
|---|---|---|
| REAG*MVFramework=QT2024.10 | 87.491 | |
| QTAugmentation Strategy=REAG_MV2024.10 | 87.491 | |
| 1T10SFramework=QT2024.10 | 87.292 | |
| QTAugmentation Strategy=1T10S2024.10 | 87.292 | |
| REAG*DaraFramework=QT2024.10 | 76.169 | |
| QTAugmentation Strategy=REAG_Dara2024.10 | 76.169 | |
| REAG*MVFramework=DT2024.10 | 73.708 | |
| DTAugmentation Strategy=REAG_MV2024.10 | 73.708 | |
| 1T10SFramework=DT2024.10 | 73.664 | |
| DTAugmentation Strategy=1T10S2024.10 | 73.664 | |
| REAG*DaraFramework=DT2024.10 | 67.565 | |
| DTAugmentation Strategy=REAG_Dara2024.10 | 67.565 | |
| 1T10SFramework=Reinformer2024.10 | 67.032 | |
| ReinformerAugmentation Strategy=1T10S2024.10 | 67.032 | |
| REAG*DaraFramework=Reinformer2024.10 | 66.658 | |
| ReinformerAugmentation Strategy=REAG_Dara2024.10 | 66.658 | |
| REAG*MVFramework=Reinformer2024.10 | 50.296 | |
| ReinformerAugmentation Strategy=REAG_MV2024.10 | 50.296 |