Offline Reinforcement Learning on MuJoCo walker2d medium-expert D4RL
116.6Normalized ReturnQCS-R
Evaluation results
| Method | Links | |
|---|---|---|
| QCS-RMethod Category=Ours2024.02 | 116.6 | |
| CGDTMethod Category=Combined Method2024.02 | 113.3 | |
| One-RL2024.10 | 113 | |
| ReBRAC2024.10 | 112.3 | |
| TD3+BCMethod Category=Value-Based2024.02 | 110.1 | |
| TD3+BC2024.10 | 110.1 | |
| SQLMethod Category=Value-Based2024.02 | 110 | |
| IQLMethod Category=Value-Based2024.02 | 109.6 | |
| DCMethod Category=RCSL2024.02 | 109.6 | |
| IQL2024.10 | 109.6 | |
| EDTMethod Category=Combined Method2024.02 | 109.3 | |
| OTR+IQL2024.10 | 109.3 | |
| ACTMethod Category=Combined Method2024.02 | 109.1 | |
| CQLMethod Category=Value-Based2024.02 | 108.8 | |
| CQL2024.10 | 108.8 | |
| DTMethod Category=RCSL2024.02 | 108.1 | |
| BC2024.10 | 107.5 | |
| RvS-RMethod Category=RCSL2024.02 | 106 |