Goal-Conditioned Reinforcement Learning on OGBench visual-humanoidmaze-medium-navigate v0
1.5Success RateOTA
Evaluation Results
| Method | Links | |
|---|---|---|
| OTAMode=Base Learner2025.10 | 1.5 | |
| OTA +TTGSMode=TTGS Augmented2025.10 | 0.9 | |
| HIQLMode=Base Learner2025.10 | 0.8 | |
| HIQL +TTGSMode=TTGS Augmented2025.10 | 0.4 | |
| SAW +TTGSMode=TTGS Augmented2025.10 | 0.2 | |
| SAWMode=Base Learner2025.10 | 0.1 | |
| GCIQLMode=Base Learner2025.10 | 0 | |
| GCIQL +TTGSMode=TTGS Augmented2025.10 | 0 | |
| QRLMode=Base Learner2025.10 | 0 | |
| QRL +TTGSMode=TTGS Augmented2025.10 | 0 |