Offline Reinforcement Learning on MuJoCo halfcheetah-medium D4RL
65.6Normalized ReturnReBRAC
Evaluation results
| Method | Links | |
|---|---|---|
| ReBRAC2024.10 | 65.6 | |
| QCS-RMethod Category=Ours2024.02 | 59 | |
| ACTMethod Category=Combined Method2024.02 | 49.1 | |
| PORMethod Category=Combined Method2024.02 | 48.8 | |
| One-RL2024.10 | 48.4 | |
| TD3+BCMethod Category=Value-Based2024.02 | 48.3 | |
| SQLMethod Category=Value-Based2024.02 | 48.3 | |
| TD3+BC2024.10 | 48.3 | |
| IQLMethod Category=Value-Based2024.02 | 47.4 | |
| IQL2024.10 | 47.4 | |
| CQLMethod Category=Value-Based2024.02 | 44 | |
| CQL2024.10 | 44 | |
| OTR+IQL2024.10 | 43.3 | |
| DCMethod Category=RCSL2024.02 | 43 | |
| CGDTMethod Category=Combined Method2024.02 | 43 | |
| DTMethod Category=RCSL2024.02 | 42.6 | |
| BC2024.10 | 42.6 | |
| EDTMethod Category=Combined Method2024.02 | 42.5 | |
| QDTMethod Category=Combined Method2024.02 | 42.3 | |
| RvS-RMethod Category=RCSL2024.02 | 41.6 |