Online Reinforcement Learning on Antmaze umaze
99.9ScoreRLPD
Evaluation Results
| Method | Links | |
|---|---|---|
| RLPDReward Setting=with true reward, Environment Steps=500k2023.10 | 99.9 | |
| BC-PEXEnvironment Steps=500k2023.10 | 99.8 | |
| UBEREnvironment Steps=500k2023.10 | 97.9 | |
| BC-PEXEnvironment Steps=100k2023.10 | 95.8 | |
| UBEREnvironment Steps=100k2023.10 | 90.7 | |
| RLPDReward Setting=with true reward, Environment Steps=100k2023.10 | 82.5 | |
| RLPDReward Setting=without true reward, Environment Steps=500k2023.10 | 0 | |
| RLPDReward Setting=without true reward, Environment Steps=100k2023.10 | 0 |