Offline Reinforcement Learning on AntMaze umaze-diverse (u-d)
84Normalized ScoreCQL
Evaluation Results
| Method | Links | |
|---|---|---|
| CQLAlgorithm Category=Value-Based Method2024.02 | 84 | |
| QCS-GAlgorithm Category=Ours, Conditioning Strategy=Goal2024.02 | 82.5 | |
| DCAlgorithm Category=RCSL2024.02 | 78.5 | |
| SQLAlgorithm Category=Value-Based Method2024.02 | 74 | |
| QCS-RAlgorithm Category=Ours, Conditioning Strategy=Return2024.02 | 72.3 | |
| TD3+BCAlgorithm Category=Value-Based Method2024.02 | 71.4 | |
| PORAlgorithm Category=Combined2024.02 | 71.3 | |
| RvS-RAlgorithm Category=RCSL, Conditioning Strategy=Return2024.02 | 70.1 | |
| IQLAlgorithm Category=Value-Based Method2024.02 | 62.2 | |
| RvS-GAlgorithm Category=RCSL, Conditioning Strategy=Goal2024.02 | 60.9 | |
| DTAlgorithm Category=RCSL2024.02 | 51.2 |