Offline Reinforcement Learning on VD4RL Cheetah-run pixel-based (medium-replay)
61.6Normalized ScoreOffline DV2
Evaluation Results
| Method | Links | |
|---|---|---|
| Offline DV2Implementation Source=Original Paper Results [20]2024.11 | 61.6 | |
| Offline DV2Implementation Source=This Paper Re-implementation2024.11 | 54.6 | |
| C-LAPImplementation Source=This Paper2024.11 | 52.5 | |
| DrQ+BCAugmentation=None2024.05 | 44.8 | |
| LOMPOImplementation Source=This Paper Re-implementation2024.11 | 42.2 | |
| GTABase Algorithm=DrQ+BC, Augmentation=GTA2024.05 | 38.1 | |
| LOMPOImplementation Source=Original Paper Results [20]2024.11 | 36.3 | |
| DrQ+BCAugmentation=SynthER2024.05 | 25.6 |