Offline Reinforcement Learning on MuJoCo halfcheetah-medium-expert D4RL
101.1Normalized ReturnReBRAC
Evaluation results
| Method | Links | |
|---|---|---|
| ReBRAC2024.10 | 101.1 | |
| ACTMethod Category=Combined Method2024.02 | 96.1 | |
| PORMethod Category=Combined Method2024.02 | 94.7 | |
| SQLMethod Category=Value-Based2024.02 | 94 | |
| CGDTMethod Category=Combined Method2024.02 | 93.6 | |
| One-RL2024.10 | 93.4 | |
| QCS-RMethod Category=Ours2024.02 | 93.3 | |
| DCMethod Category=RCSL2024.02 | 93 | |
| RvS-RMethod Category=RCSL2024.02 | 92.2 | |
| CQLMethod Category=Value-Based2024.02 | 91.6 | |
| CQL2024.10 | 91.6 | |
| TD3+BCMethod Category=Value-Based2024.02 | 90.7 | |
| TD3+BC2024.10 | 90.7 | |
| OTR+IQL2024.10 | 89.6 | |
| DTMethod Category=RCSL2024.02 | 86.8 | |
| IQLMethod Category=Value-Based2024.02 | 86.7 | |
| IQL2024.10 | 86.7 | |
| BC2024.10 | 55.2 |