Offline Reinforcement Learning on D4RL MuJoCo walker2d-medium
91.7Normalized ScoreBFQ
Evaluation results
| Method | Links | |
|---|---|---|
| BFQPolicy Type=One-Step Flow/Diffusion Policy2026.06 | 91.7 | |
| MCQPolicy Type=Gaussian, Seeds=82026.02 | 91 | |
| ISEP-FMFlow Matching=true2026.05 | 90.4 | |
| EPQPolicy Type=Gaussian, Seeds=82026.02 | 87.8 | |
| QIPO-OTPolicy Type=Diffusion, Seeds=82026.02 | 87.6 | |
| QIPO2026.05 | 87.6 | |
| OFQLPolicy Type=One-Step Flow/Diffusion Policy2026.06 | 87.4 | |
| DQL2026.05 | 87 | |
| DQLPolicy Type=Diffusion Policy2026.06 | 86.8 | |
| EDPPolicy Type=Diffusion Policy2026.06 | 86.5 | |
| SPOTPolicy Type=Gaussian, Seeds=82026.02 | 86.4 | |
| IDQLPolicy Type=Diffusion Policy2026.06 | 86.1 | |
| QGPO2026.05 | 86 | |
| FACPolicy Type=Flow, Seeds=82026.02 | 85.2 | |
| SRPOPolicy Type=Diffusion, Seeds=82026.02 | 84.4 | |
| SRPOPolicy Type=One-Step Flow/Diffusion Policy2026.06 | 84.4 | |
| TD3+BCPolicy Type=Gaussian, Seeds=82026.02 | 83.7 | |
| TD3-BCPolicy Type=Gaussian Policy2026.06 | 83.3 | |
| CACPolicy Type=Diffusion, Seeds=82026.02 | 83.1 | |
| ReBRACPolicy Type=Gaussian, Seeds=82026.02 | 82.5 | |
| IDQLPolicy Type=Diffusion, Seeds=82026.02 | 82.5 | |
| ISEPFlow Matching=false2026.05 | 82.1 | |
| IDQL2026.05 | 80.2 | |
| IQLPolicy Type=Gaussian Policy2026.06 | 78.7 | |
| IQLPolicy Type=Gaussian, Seeds=82026.02 | 78.3 | |
| IQL2026.05 | 78.3 | |
| SORLPolicy Type=One-Step Flow/Diffusion Policy2026.06 | 77.9 | |
| FQLPolicy Type=Flow, Seeds=82026.02 | 77.2 | |
| BC2026.05 | 75.3 | |
| FQLPolicy Type=One-Step Flow/Diffusion Policy2026.06 | 72.7 | |
| CQLPolicy Type=Gaussian, Seeds=82026.02 | 72.5 | |
| BCPolicy Type=Gaussian Policy2026.06 | 72.3 | |
| CQL2026.05 | 58 |