Offline Reinforcement Learning on DeepMind Control Suite cheetah-run (medium)
62.6Mean Normalized ReturnTD3+BC
Evaluation Results
| Method | Links | |
|---|---|---|
| TD3+BCaction-conditioned data percentage=100%, variant=Oracle2025.12 | 62.6 | |
| IDM-TD3+BCaction-conditioned data percentage=5%2025.12 | 61.3 | |
| C-LAPaction-conditioned data percentage=100%, variant=Oracle2025.12 | 57.8 | |
| LAWMaction-conditioned data percentage=5%2025.12 | 54.6 | |
| C-LAPaction-conditioned data percentage=5%2025.12 | 47.9 | |
| TD3+BCaction-conditioned data percentage=5%2025.12 | 32.4 |