Offline Reinforcement Learning on OGBench humanoidmaze-medium-stitch-singletask-task1
47Test ReturnBiTrajDiff
Evaluation Results
| Method | Links | |
|---|---|---|
| BiTrajDiffBase RL Algorithm=FQL, Data Augmentation Method=Ours2025.06 | 47 | |
| SSDBase RL Algorithm=FQL, Data Augmentation Method=SSD2025.06 | 43 | |
| BiTrajDiffBase RL Algorithm=IQL, Data Augmentation Method=Ours2025.06 | 38 | |
| SSDBase RL Algorithm=IQL, Data Augmentation Method=SSD2025.06 | 33 | |
| ASTROBase RL Algorithm=IQL, Data Augmentation Method=ASTRO2025.06 | 31.4 | |
| SyntherBase RL Algorithm=IQL, Data Augmentation Method=Synther2025.06 | 31.2 | |
| ASTROBase RL Algorithm=FQL, Data Augmentation Method=ASTRO2025.06 | 30 | |
| IQLBase RL Algorithm=IQL, Data Augmentation Method=Base2025.06 | 29.7 | |
| DiffStitchBase RL Algorithm=IQL, Data Augmentation Method=DiffStitch2025.06 | 28.3 | |
| DiffStitchBase RL Algorithm=FQL, Data Augmentation Method=DiffStitch2025.06 | 22.6 | |
| FQLBase RL Algorithm=FQL, Data Augmentation Method=Base2025.06 | 17.5 | |
| SyntherBase RL Algorithm=FQL, Data Augmentation Method=Synther2025.06 | 15.9 |