Goal-conditioned Offline Reinforcement Learning on OGBench antmaze-medium-stitch v0
68ReturnBiTrajDiff
Evaluation Results
| Method | Links | |
|---|---|---|
| BiTrajDiffRL Algorithm=CRL2025.06 | 68 | |
| SCoTSRL Algorithm=CRL, Augmentation=SCoTS2025.06 | 65 | |
| CDRL Algorithm=CRL, Augmentation=CD2025.06 | 61 | |
| CRLRL Algorithm=CRL, Augmentation=None (Base)2025.06 | 53 | |
| SyntherRL Algorithm=CRL, Augmentation=Synther2025.06 | 48 | |
| BiTrajDiffRL Algorithm=GCIQL2025.06 | 38 | |
| SCoTSRL Algorithm=GCIQL, Augmentation=SCoTS2025.06 | 35 | |
| CDRL Algorithm=GCIQL, Augmentation=CD2025.06 | 34 | |
| SyntherRL Algorithm=GCIQL, Augmentation=Synther2025.06 | 31 | |
| GCIQLRL Algorithm=GCIQL, Augmentation=None (Base)2025.06 | 29 |