Offline Reinforcement Learning on DeepMind Control Suite hopper-stand medium-replay
63.1Mean Normalized ReturnC-LAP
Evaluation Results
| Method | Links | |
|---|---|---|
| C-LAPaction-conditioned data percentage=100%, variant=Oracle2025.12 | 63.1 | |
| LAWMaction-conditioned data percentage=5%2025.12 | 46.9 | |
| C-LAPaction-conditioned data percentage=5%2025.12 | 19 | |
| TD3+BCaction-conditioned data percentage=100%, variant=Oracle2025.12 | 15.3 | |
| IDM-TD3+BCaction-conditioned data percentage=5%2025.12 | 13.7 | |
| TD3+BCaction-conditioned data percentage=5%2025.12 | 10.2 |