Walker2d Locomotion Control on D4RL walker2d medium-replay
106.2Normalized ScoreOFQL
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| OFQLCategory=One-step Flow Policies2025.08 | 106.2 | — | — | |
| DQLCategory=Multi-step Diffusion Policies2025.08 | 95.5 | — | — | |
| BestPhase=Offline2026.05 | 94.9 | — | — | |
| OEPhase=Online, Online Interaction Budget=320 K2026.05 | 94.9 | — | — | |
| Adaptive Policy Selection and Fine-TuningPhase=Online, Online Interaction Budget=320 K2026.05 | 92.1 | — | — | |
| PRDC2024.03 | 92 | — | — | |
| IDQLCategory=Multi-step Diffusion Policies2025.08 | 89.1 | — | — | |
| EDPCategory=Multi-step Diffusion Policies2025.08 | 87 | — | — | |
| DMEMM2026.02 | 85.8 | — | — | |
| SyntherRL Algorithm=CQL2025.06 | 84.8 | — | — | |
| BiTrajDiffRL Algorithm=CQL2025.06 | 84.8 | — | — | |
| RTDiffRL Algorithm=CQL2025.06 | 84.4 | — | — | |
| HD-DA2026.02 | 84.1 | — | — | |
| A2PO2024.03 | 82.8 | — | — | |
| TT2023.12 | 82.6 | — | — | |
| TT2026.02 | 82.6 | — | — | |
| TT2026.03 | 82.6 | — | — | |
| DiffStitchRL Algorithm=CQL2025.06 | 82.4 | — | — | |
| FQLCategory=One-step Flow Policies2025.08 | 82.1 | — | — | |
| BaseRL Algorithm=CQL2025.06 | 82.1 | — | — | |
| TD3+BCMethod Category=Best RL Baseline, Seeds=52022.10 | 81.8 | — | — | |
| TD3-BCCategory=Non-Diffusion Policies2025.08 | 81.8 | — | — | |
| HDMI2026.02 | 80.7 | — | — | |
| DTAMP2023.12 | 79.5 | — | — | |
| IQL+ReD2024.03 | 79.5 | — | — | |
| CQL2026.02 | 77.2 | — | — | |
| CQL2026.02 | 77.2 | — | — | |
| CQL2026.03 | 77.2 | — | — | |
| PORMethod Category=Conditioned BC, Seeds=52022.10 | 76.6 | — | — | |
| DDCategory=Diffusion Planners2025.08 | 75.6 | — | — | |
| DD2023.12 | 75 | — | — | |
| IQL2023.12 | 73.9 | — | — | |
| IQL2026.02 | 73.9 | — | — | |
| IQLCategory=Non-Diffusion Policies2025.08 | 73.9 | — | — | |
| IQL2026.02 | 73.9 | — | — | |
| IQL2026.03 | 73.9 | — | — | |
| IQLMethod Category=Weighted BC, Seeds=52022.10 | 73.8 | — | — | |
| Diffuser2026.02 | 70.6 | — | — | |
| Diffuser2026.03 | 70.6 | — | — | |
| IMLE+Exp RW2026.03 | 69.7 | — | — | |
| ISCTSeeds=5, Episodes=102026.02 | 68.9 | — | — | |
| DTMethod Category=Conditioned BC, Seeds=52022.10 | 66.6 | — | — | |
| DT2023.12 | 66.6 | — | — | |
| DT2024.03 | 66.6 | — | — | |
| DT2026.02 | 66.6 | — | — | |
| DT2026.02 | 66.6 | — | — | |
| DT2026.03 | 66.6 | — | — | |
| 10%BCMethod Category=Weighted BC, Seeds=52022.10 | 62.5 | — | — | |
| %BC2024.03 | 62.5 | — | — | |
| Diffuser2023.12 | 61.2 | — | — | |
| RvSMethod Category=Conditioned BC, Seeds=52022.10 | 60.6 | — | — | |
| DiffuserCategory=Diffusion Planners2025.08 | 58.3 | — | — | |
| BiTrajDiffRL Algorithm=DT2025.06 | 58.2 | — | — | |
| FTPhase=Online, Online Interaction Budget=320 K2026.05 | 56.4 | — | — | |
| SyntherRL Algorithm=DT2025.06 | 56.1 | — | — | |
| COMBO2021.02 | 56 | — | — | |
| DiffStitchRL Algorithm=DT2025.06 | 54.6 | — | — | |
| RTDiffRL Algorithm=DT2025.06 | 53.8 | — | — | |
| BaseRL Algorithm=DT2025.06 | 52.1 | — | — | |
| MOREL2021.02 | 49.8 | — | — | |
| MOReL2026.02 | 49.8 | — | — | |
| MOReL2026.03 | 49.8 | — | — | |
| One-step RLMethod Category=Weighted BC, Seeds=52022.10 | 49.5 | — | — | |
| MOPO2021.02 | 39 | — | — | |
| MOPO2026.03 | 39 | — | — | |
| OPEPhase=Offline2026.05 | 33.8 | — | — | |
| CQL2021.02 | 32.6 | — | — | |
| CQL2023.12 | 26 | — | — | |
| BCCategory=Non-Diffusion Policies2025.08 | 26 | — | — | |
| BC2026.02 | 26 | — | — | |
| BC2026.03 | 26 | — | — | |
| TD3+BC2026.02 | 25.2 | — | — | |
| BEAR2021.02 | 19.2 | — | — | |
| BC2021.02 | 11.3 | — | — | |
| MBOP2026.03 | 9.7 | — | — | |
| SAC-off2021.02 | 1.9 | — | — | |
| BRAC-v2021.02 | 0.9 | — | — | |
| BRAC-p2021.02 | -0.3 | — | — | |
| O2PPOOnline update way=IQL2024.12 | — | 90.43 | 11.6 | |
| O2SACOnline update way=SAC2024.12 | — | 106.05 | 28.75 | |
| Off2OnOnline update way=SAC2024.12 | — | 102.43 | 20.78 |