Goal-conditioned Reinforcement Learning on pointmaze navigate medium
99Success RateEik-HiQRL
Evaluation Results
| Method | Links | |
|---|---|---|
| Eik-HiQRL2025.12 | 99 | |
| Eik-HIQL2025.12 | 93 | |
| HIQL2025.12 | 92 | |
| QRL2025.12 | 83 | |
| OrigDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 78 | |
| Dual Goal RepresentationsDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 76 | |
| VIBDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 69 | |
| CRL2025.12 | 54 | |
| BYOL-γDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 37 | |
| TRADownstream algorithm=GCIVL, Observation type=State-based2025.10 | 3 | |
| VIPDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 0 |