Offline Reinforcement Learning on D4RL Hopper-medium-replay v2
110.7Normalized ReturnPhyB
Evaluation Results
| Method | Links | |
|---|---|---|
| PhyBModel category=Model-based2026.05 | 110.7 | |
| EDAC2022.10 | 107.4 | |
| PMDB2022.10 | 106.2 | |
| PMDBModel category=Model-based2026.05 | 106.2 | |
| ADMModel category=Model-based2026.05 | 104.4 | |
| Uncertainty-Aware Rank-One MIMO Q Network2026.02 | 102.9 | |
| EDAC-102026.02 | 102.8 | |
| DMGModel category=Model-free2026.05 | 101.9 | |
| PBRL2026.02 | 100.6 | |
| SPOT2022.02 | 100.2 | |
| SPOT2023.01 | 100.2 | |
| RAMBO2022.10 | 99.5 | |
| RAMBOModel category=Model-based2026.05 | 99.5 | |
| CPED2023.01 | 98.1 | |
| EPQModel category=Model-free2026.05 | 97.8 | |
| Onestep2022.02 | 97.5 | |
| Onestep RL2023.01 | 97.5 | |
| CQL2022.02 | 95 | |
| CQL2023.01 | 95 | |
| IQL2022.02 | 94.7 | |
| IQL2023.01 | 94.7 | |
| MOReL2023.11 | 93.6 | |
| CQL2026.02 | 88.7 | |
| CQL2022.10 | 86.3 | |
| CQLModel category=Model-free2026.05 | 86.3 | |
| TAILO2023.11 | 83.4 | |
| DT2022.02 | 82.7 | |
| DT2023.01 | 82.7 | |
| IQL2026.02 | 81.9 | |
| MOREL2022.10 | 81.8 | |
| MOReLModel category=Model-based2026.05 | 81.8 | |
| GACStrategy=Exploitation query (GAC-E[y])2025.12 | 80.1 | |
| LPT2025.12 | 71.2 | |
| MOPO2023.11 | 67.5 | |
| GACStrategy=Exploration query (GAC-p(y+))2025.12 | 67.5 | |
| BRAC2022.10 | 62.7 | |
| GACStrategy=Fixed target steering (GAC-y*)2025.12 | 61.3 | |
| TD3+BC2022.02 | 60.9 | |
| TD3+BC2023.01 | 60.9 | |
| TD3-BC2026.02 | 60.9 | |
| BEAR2022.10 | 57.7 | |
| BEAR2026.02 | 52.2 | |
| DT2025.12 | 50.8 | |
| DROCOShift Type=Kinematic2025.12 | 47.3 | |
| CQLShift Type=Kinematic2025.12 | 43.2 | |
| DARAShift Type=Kinematic2025.12 | 41.6 | |
| FQLModel category=Model-free2026.05 | 40.6 | |
| IQLShift Type=Kinematic2025.12 | 40.2 | |
| QDT2025.12 | 38.7 | |
| AWAC2022.02 | 37.2 | |
| AWAC2023.01 | 37.2 | |
| GACStrategy=Sampling from prior (p(y|z)p(z))2025.12 | 34.3 | |
| MOPO2026.02 | 32.7 | |
| TD3+BCModel category=Model-free2026.05 | 31.4 | |
| IGDFShift Type=Kinematic2025.12 | 30 | |
| BOSAShift Type=Kinematic2025.12 | 29.5 | |
| OTDFShift Type=Kinematic2025.12 | 26.2 | |
| UWAV2026.02 | 25.3 | |
| BC2022.02 | 18.1 | |
| BC2023.01 | 18.1 | |
| BC2022.10 | 16.6 | |
| IQL2025.12 | 13.9 | |
| BCQ2026.02 | 11.1 | |
| CQL2025.12 | 7.7 |