Locomotion on D4RL HalfCheetah (medium)
63.5Normalized ScorePRDC
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| PRDC2024.03 | 63.5 | — | — | |
| SAGE2026.03 | 51.6 | — | — | |
| DVreproduced=true2026.03 | 50.4 | — | — | |
| DMEMM2026.02 | 49.2 | — | — | |
| DD2023.12 | 49.1 | — | — | |
| PORMethod Category=Conditioned BC, Seeds=52022.10 | 48.8 | — | — | |
| BiTrajDiffRL Algorithm=CQL2025.06 | 48.7 | — | — | |
| One-step RLMethod Category=Weighted BC, Seeds=52022.10 | 48.4 | — | — | |
| Onestep RLTraining Time (min)=202023.06 | 48.4 | — | — | |
| TD3+BCMethod Category=Best RL Baseline, Seeds=52022.10 | 48.3 | — | — | |
| TD3 + BCTraining Time (min)=202023.06 | 48.3 | — | — | |
| SyntherRL Algorithm=CQL2025.06 | 48.3 | — | — | |
| HDMI2026.02 | 48 | — | — | |
| DiffStitchRL Algorithm=CQL2025.06 | 48 | — | — | |
| RTDiffRL Algorithm=CQL2025.06 | 47.8 | — | — | |
| IQL+ReD2024.03 | 47.6 | — | — | |
| IQLMethod Category=Weighted BC, Seeds=52022.10 | 47.4 | — | — | |
| IQLTraining Time (min)=202023.06 | 47.4 | — | — | |
| IQL2023.12 | 47.4 | — | — | |
| IQL2026.02 | 47.4 | — | — | |
| IQL2026.02 | 47.4 | — | — | |
| IQL2026.03 | 47.4 | — | — | |
| BaseRL Algorithm=CQL2025.06 | 47.4 | — | — | |
| DTAMP2023.12 | 47.3 | — | — | |
| A2PO2024.03 | 47.1 | — | — | |
| TT2023.12 | 46.9 | — | — | |
| TT2026.02 | 46.9 | — | — | |
| TT2026.03 | 46.9 | — | — | |
| HD-DA2026.02 | 46.7 | — | — | |
| LoMAP2026.03 | 45.4 | — | — | |
| MBOP2026.03 | 44.6 | — | — | |
| Diffuser2023.12 | 44.2 | — | — | |
| CQLTraining Time (min)=802023.06 | 44 | — | — | |
| CQL2026.02 | 44 | — | — | |
| CQL2026.02 | 44 | — | — | |
| RGG2026.03 | 44 | — | — | |
| CQL2026.03 | 44 | — | — | |
| BiTrajDiffRL Algorithm=DT2025.06 | 43.3 | — | — | |
| IMLE+Exp RW2026.03 | 43.1 | — | — | |
| WT (Ours)Training Time (min)=202023.06 | 43 | — | — | |
| ISCTSeeds=5, Episodes=102026.02 | 42.9 | — | — | |
| TD3+BC2026.02 | 42.8 | — | — | |
| Diffuser2026.02 | 42.8 | — | — | |
| Diffuserreproduced=true2026.03 | 42.8 | — | — | |
| LDCQ2026.03 | 42.8 | — | — | |
| Diffuser2026.03 | 42.8 | — | — | |
| DiffStitchRL Algorithm=DT2025.06 | 42.8 | — | — | |
| BaseRL Algorithm=DT2025.06 | 42.7 | — | — | |
| SyntherRL Algorithm=DT2025.06 | 42.7 | — | — | |
| DTMethod Category=Conditioned BC, Seeds=52022.10 | 42.6 | — | — | |
| BC 10%Training Time (min)=102023.06 | 42.6 | — | — | |
| CQL2023.12 | 42.6 | — | — | |
| DT2023.12 | 42.6 | — | — | |
| DT2024.03 | 42.6 | — | — | |
| DT2026.02 | 42.6 | — | — | |
| BC2026.02 | 42.6 | — | — | |
| DT2026.02 | 42.6 | — | — | |
| BC2026.03 | 42.6 | — | — | |
| DT2026.03 | 42.6 | — | — | |
| RTDiffRL Algorithm=DT2025.06 | 42.6 | — | — | |
| 10%BCMethod Category=Weighted BC, Seeds=52022.10 | 42.5 | — | — | |
| BCTraining Time (min)=102023.06 | 42.5 | — | — | |
| %BC2024.03 | 42.5 | — | — | |
| DTTraining Time (min)=1502023.06 | 42.4 | — | — | |
| QDT2023.06 | 42.3 | — | — | |
| MOPO2026.03 | 42.3 | — | — | |
| MOReL2026.02 | 42.1 | — | — | |
| MOReL2026.03 | 42.1 | — | — | |
| RvSMethod Category=Conditioned BC, Seeds=52022.10 | 41.6 | — | — | |
| RvS-R/GTraining Time (min)=802023.06 | 41.6 | — | — | |
| ACAOnline update way=SAC, Evaluation protocol=Offline-to-online2024.12 | — | 66.59 | 20.09 | |
| AWACOnline update way=IQL/Policy constraint, Evaluation protocol=Offline-to-online2024.12 | — | 51.11 | 1.43 | |
| Cal-QLOnline update way=SAC, Evaluation protocol=Offline-to-online2024.12 | — | 54.36 | 6.81 | |
| IQLOnline update way=IQL/Policy constraint, Evaluation protocol=Offline-to-online2024.12 | — | 48.85 | 0.33 | |
| O2PPOOnline update way=IQL/Policy constraint, Evaluation protocol=Offline-to-online2024.12 | — | 59.96 | 11.54 | |
| O2SACOnline update way=SAC, Evaluation protocol=Offline-to-online2024.12 | — | 74.18 | 27.32 | |
| O2TD3Online update way=TD3, Evaluation protocol=Offline-to-online2024.12 | — | 65.58 | 17.45 | |
| Off2OnOnline update way=SAC, Evaluation protocol=Offline-to-online2024.12 | — | 94.22 | 47.37 | |
| PEXOnline update way=IQL/Policy constraint, Evaluation protocol=Offline-to-online2024.12 | — | 57.05 | 8.53 |