Goal-Conditioned Reinforcement Learning on cube-double-play state-based v0 (test)
60Success RateDual Goal Representations
Evaluation Results
| Method | Links | |
|---|---|---|
| Dual Goal RepresentationsDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 60 | |
| OrigDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 35 | |
| VIBDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 33 | |
| TRADownstream algorithm=GCIVL, Observation type=State-based2025.10 | 7 | |
| BYOL-γDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 6 | |
| VIPDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 3 |