Offline Reinforcement Learning on D4RL walker2d-expert v2
116.3Normalized ScorePhyB
Evaluation Results
| Method | Links | |
|---|---|---|
| PhyBModel category=Model-based2026.05 | 116.3 | |
| PMDB2022.10 | 115.9 | |
| PMDBN=10, k=22022.10 | 115.9 | |
| PMDBModel category=Model-based2026.05 | 115.9 | |
| CEILProtocol=S-off-LfD, Trials=302023.06 | 115.6 | |
| EDAC2022.10 | 115.1 | |
| EDAC2022.10 | 115.1 | |
| DMGModel category=Model-free2026.05 | 114.7 | |
| Uncertainty-Aware Rank-One MIMO Q Network2026.02 | 112.7 | |
| BCQ2026.02 | 110.4 | |
| TD3-BC2026.02 | 110.2 | |
| BEAR2022.10 | 110.1 | |
| BEAR2022.10 | 110.1 | |
| CEILProtocol=S-off-LfO, Trials=302023.06 | 110 | |
| EPQModel category=Model-free2026.05 | 109.8 | |
| CQL2022.10 | 109.3 | |
| CQL2022.10 | 109.3 | |
| CQLModel category=Model-free2026.05 | 109.3 | |
| IQL2026.02 | 109 | |
| BC2022.10 | 108.7 | |
| BC2022.10 | 108.7 | |
| CEILProtocol=C-off-LfO, Trials=302023.06 | 108.7 | |
| CEILProtocol=C-off-LfD, Trials=302023.06 | 108.6 | |
| CQL2026.02 | 108.5 | |
| UWAV2026.02 | 108.4 | |
| Decision Mamba2024.06 | 108.3 | |
| PBRL2026.02 | 108.3 | |
| SMODICEProtocol=S-off-LfD, Trials=302023.06 | 108.2 | |
| SMODICEProtocol=C-off-LfD, Trials=302023.06 | 108.2 | |
| SMODICEProtocol=C-off-LfO, Trials=302023.06 | 108.1 | |
| DT2024.06 | 108.1 | |
| LaMo2024.06 | 108.1 | |
| SMODICEProtocol=S-off-LfO, Trials=302023.06 | 108 | |
| DemoDICEProtocol=C-off-LfD, Trials=302023.06 | 107.9 | |
| BEAR2026.02 | 106.6 | |
| DROCOShift Type=Kinematic2025.12 | 106 | |
| TD3+BCModel category=Model-free2026.05 | 105.7 | |
| DemoDICEProtocol=S-off-LfD, Trials=302023.06 | 104.8 | |
| BC2024.06 | 103.2 | |
| OTDFShift Type=Kinematic2025.12 | 98.9 | |
| IGDFShift Type=Kinematic2025.12 | 93.7 | |
| IQLShift Type=Kinematic2025.12 | 90.1 | |
| DARAShift Type=Kinematic2025.12 | 85.5 | |
| CQLShift Type=Kinematic2025.12 | 83.8 | |
| MOREL2022.10 | 62.6 | |
| MOReL2022.10 | 62.6 | |
| MOReLModel category=Model-based2026.05 | 62.6 | |
| MOPO2026.02 | 62.4 | |
| EDAC-102026.02 | 57.8 | |
| BRAC2022.10 | 55.2 | |
| BRAC2022.10 | 55.2 | |
| IQ-LearnProtocol=S-off-LfD, Trials=302023.06 | 46.6 | |
| BOSAShift Type=Kinematic2025.12 | 41.9 | |
| IQ-LearnProtocol=C-off-LfD, Trials=302023.06 | 31.7 | |
| ORIL (TD3+BC)Protocol=C-off-LfD, Trials=302023.06 | 29.3 | |
| ValueDICEProtocol=S-off-LfD, Trials=302023.06 | 28.2 | |
| ORIL (TD3+BC)Protocol=S-off-LfD, Trials=302023.06 | 14.5 | |
| FQLModel category=Model-free2026.05 | 12.7 | |
| ORIL (TD3+BC)Protocol=S-off-LfO, Trials=302023.06 | 12.2 | |
| ORIL (TD3+BC)Protocol=C-off-LfO, Trials=302023.06 | 11.2 | |
| SQIL (TD3+BC)Protocol=C-off-LfD, Trials=302023.06 | 8.8 | |
| ValueDICEProtocol=C-off-LfD, Trials=302023.06 | 8.3 | |
| SQIL (TD3+BC)Protocol=S-off-LfD, Trials=302023.06 | 8 | |
| ADMModel category=Model-based2026.05 | 5.5 | |
| RAMBO2022.10 | 1.6 | |
| RAMBOModel category=Model-based2026.05 | 1.6 |