Offline Reinforcement Learning on AntMaze medium-diverse (m-d)
79.5Normalized ScoreQCS-R
Evaluation Results
| Method | Links | |
|---|---|---|
| QCS-RAlgorithm Category=Ours, Conditioning Strategy=Return2024.02 | 79.5 | |
| PORAlgorithm Category=Combined2024.02 | 79.2 | |
| SQLAlgorithm Category=Value-Based Method2024.02 | 79.1 | |
| QCS-GAlgorithm Category=Ours, Conditioning Strategy=Goal2024.02 | 75.2 | |
| IQLAlgorithm Category=Value-Based Method2024.02 | 70 | |
| RvS-GAlgorithm Category=RCSL, Conditioning Strategy=Goal2024.02 | 67.3 | |
| CQLAlgorithm Category=Value-Based Method2024.02 | 53.7 | |
| DCAlgorithm Category=RCSL2024.02 | 27.5 | |
| RvS-RAlgorithm Category=RCSL, Conditioning Strategy=Return2024.02 | 7.7 | |
| TD3+BCAlgorithm Category=Value-Based Method2024.02 | 3 | |
| DTAlgorithm Category=RCSL2024.02 | 1.2 |