Offline Reinforcement Learning on AntMaze umaze (Normalized Score)
92.7Normalized ScoreQCS-R
Evaluation Results
| Method | Links | |
|---|---|---|
| QCS-RAlgorithm Category=Ours, Conditioning Strategy=Return2024.02 | 92.7 | |
| QCS-GAlgorithm Category=Ours, Conditioning Strategy=Goal2024.02 | 92.5 | |
| SQLAlgorithm Category=Value-Based Method2024.02 | 92.2 | |
| PORAlgorithm Category=Combined2024.02 | 90.6 | |
| IQLAlgorithm Category=Value-Based Method2024.02 | 87.5 | |
| DCAlgorithm Category=RCSL2024.02 | 85 | |
| TD3+BCAlgorithm Category=Value-Based Method2024.02 | 78.6 | |
| CQLAlgorithm Category=Value-Based Method2024.02 | 74 | |
| DTAlgorithm Category=RCSL2024.02 | 65.6 | |
| RvS-GAlgorithm Category=RCSL, Conditioning Strategy=Goal2024.02 | 65.4 | |
| RvS-RAlgorithm Category=RCSL, Conditioning Strategy=Return2024.02 | 64.4 |