Offline Reinforcement Learning on D4RL Gym hopper-random
35.6Normalized ScoreSPQR
Evaluation Results
| Method | Links | |
|---|---|---|
| SPQRBase algorithm=SAC-Min2024.01 | 35.6 | |
| RORLImplementation Source=Ours, Ensemble Size=102022.06 | 31.4 | |
| SAC-Min2024.01 | 31.3 | |
| PBRL2022.06 | 26.8 | |
| SAC-10Implementation Source=Reproduced, Ensemble Size=102022.06 | 25.9 | |
| EDACImplementation Source=Paper2022.06 | 25.3 | |
| EDAC2024.01 | 25.3 | |
| EDAC-10Implementation Source=Reproduced, Ensemble Size=102022.06 | 16.9 | |
| CQL-Min2024.01 | 7 | |
| CQL2022.06 | 5.3 | |
| BC2022.06 | 3.7 | |
| BC2024.01 | 3.7 |