Goal-conditioned Offline Reinforcement Learning on OGBench antmaze-medium-explore v0
19ReturnBiTrajDiff
Evaluation Results
| Method | Links | |
|---|---|---|
| BiTrajDiffRL Algorithm=GCIQL2025.06 | 19 | |
| SCoTSRL Algorithm=GCIQL, Augmentation=SCoTS2025.06 | 18 | |
| CDRL Algorithm=GCIQL, Augmentation=CD2025.06 | 17 | |
| SCoTSRL Algorithm=CRL, Augmentation=SCoTS2025.06 | 15 | |
| BiTrajDiffRL Algorithm=CRL2025.06 | 14 | |
| GCIQLRL Algorithm=GCIQL, Augmentation=None (Base)2025.06 | 13 | |
| SyntherRL Algorithm=GCIQL, Augmentation=Synther2025.06 | 12 | |
| CDRL Algorithm=CRL, Augmentation=CD2025.06 | 11 | |
| CRLRL Algorithm=CRL, Augmentation=None (Base)2025.06 | 3 | |
| SyntherRL Algorithm=CRL, Augmentation=Synther2025.06 | 3 |