Goal-Conditioned Reinforcement Learning (Navigation) on AntMaze Giant Navigate (state-based v0, test)
0Success RateOrig
Evaluation results
| Method | Links | |
|---|---|---|
| OrigDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 0 | |
| VIBDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 0 | |
| VIPDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 0 | |
| TRADownstream algorithm=GCIVL, Observation type=State-based2025.10 | 0 | |
| BYOL-γDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 0 | |
| Dual Goal RepresentationsDownstream algorithm=GCIVL, Observation type=State-based2025.10 | 0 |