Offline Reinforcement Learning on D4RL walker2d-medium-replay
99.3Normalized ScoreNEUBAY
Evaluation results
| Method | Links | |
|---|---|---|
| NEUBAYMethod Category=Ours2025.12 | 99.3 | |
| LEQMethod Category=Conservative model-based2025.12 | 98.7 | |
| VIPOMethod Category=Conservative model-based2025.12 | 98.4 | |
| RRPINumber of random seeds=42026.03 | 96.4 | |
| ADMPOMethod Category=Conservative model-based2025.12 | 95.6 | |
| AMGNumber of random seeds=42026.03 | 95.6 | |
| CBOPMethod Category=Bayesian-inspired2025.12 | 92.7 | |
| ATACMethod Type=Model-free2023.10 | 92.5 | |
| CROPMethod Type=Model-based2023.10 | 91 | |
| MoMoMethod Category=Conservative model-based2025.12 | 90.4 | |
| MOBILEImplementation=Original Paper2024.11 | 89.9 | |
| MOBILEMethod Category=Conservative model-based2025.12 | 89.9 | |
| DMGNumber of random seeds=42026.03 | 89.7 | |
| RAMBONumber of random seeds=42026.03 | 89.2 | |
| MoDAPMethod Category=Bayesian-inspired2025.12 | 88.4 | |
| Count-MORLMethod Type=Model-based2023.10 | 87.7 | |
| EDACMethod Category=Model-free2025.12 | 87.1 | |
| EDACMethod Type=Model-free2023.10 | 87.1 | |
| RAMBOMethod Category=Model-free2025.12 | 86.9 | |
| C-LAPImplementation=Reproduced2024.11 | 86 | |
| MOPOMethod Category=Conservative model-based2025.12 | 85.6 | |
| MOBILEImplementation=Reproduced2024.11 | 85.4 | |
| EPQNumber of random seeds=42026.03 | 85.3 | |
| MOPOImplementation=Original Paper2024.11 | 85.2 | |
| RAMBOMethod Type=Model-based2023.10 | 85 | |
| AdaptDiffuserCategory=Diffusion Planner2024.05 | 84.4 | |
| APE-VMethod Category=Conservative model-based2025.12 | 82.9 | |
| IQLAccess=True reward functions2024.04 | 82.2 | |
| TD3+BCModel Category=Value-based2026.02 | 81.8 | |
| PMDBNumber of random seeds=42026.03 | 79.9 | |
| PMDBMethod Type=Model-based2023.10 | 79.9 | |
| SUMOMethod Category=Conservative model-based2025.12 | 78.2 | |
| CQLModel Category=Value-based2026.02 | 77.2 | |
| DCModel Category=SSM-based2026.02 | 77.1 | |
| CQLMethod Category=Model-free2025.12 | 76.8 | |
| CQLNumber of random seeds=42026.03 | 76.8 | |
| PTAccess=Scripted preferences2024.04 | 75.7 | |
| MAPLEMethod Category=Bayesian-inspired2025.12 | 75.4 | |
| DDCategory=Diffusion Planner2024.05 | 75 | |
| DMMModel Category=SSM-based2026.02 | 74.1 | |
| IQLMethod Type=Model-free2023.10 | 73.9 | |
| EDTModel Category=Transformer-based2026.02 | 73.3 | |
| DMModel Category=SSM-based2026.02 | 72.5 | |
| OPPOAccess=Scripted preferences, Implementation=Reported2024.04 | 71.7 | |
| DiffuserCategory=Diffusion Planner2024.05 | 70.6 | |
| DS4Model Category=SSM-based2026.02 | 69 | |
| PLASImplementation=Reproduced2024.11 | 67.7 | |
| DTModel Category=Transformer-based2026.02 | 65.9 | |
| CAMPAccess=Scripted preferences2024.04 | 60.5 | |
| QLDTModel Category=Transformer-based2026.02 | 58.2 | |
| COMBOCategory=Model-based RL2024.05 | 56 | |
| COMBOMethod Category=Model-free2025.12 | 56 | |
| COMBOMethod Type=Model-based2023.10 | 56 | |
| MOPOImplementation=Reproduced2024.11 | 54.6 | |
| OPRLAccess=Scripted preferences2024.04 | 53.3 | |
| MORELCategory=Model-based RL2024.05 | 49.8 | |
| MOReLNumber of random seeds=42026.03 | 40.8 | |
| MOPOCategory=Model-based RL2024.05 | 39 | |
| CABI+TD3-BCMethod Type=Model-based2023.10 | 29.4 | |
| BCAccess=Scripted preferences2024.04 | 21.8 | |
| OPPOAccess=Scripted preferences, Implementation=Re-implementation2024.04 | 19.7 | |
| PLASImplementation=Original Paper2024.11 | 10.2 |