Zero-shot Reinforcement Learning on ExORL APS (Jaco environment) v1 (test)
88Reach Bottom LeftHILP
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| HILPMode=Zero-shot, Seeds=42024.06 | 88 | 48 | 49 | 51 | |
| DoDontMode=Zero-shot, Seeds=42024.06 | 61 | 60 | 88 | 64 | |
| FDMMode=Zero-shot, Seeds=42024.06 | 53 | 38 | 36 | 44 | |
| DoDontMode=Zero-shot, Seeds=42024.06 | 33 | 29 | 33 | 9 | |
| HILPMode=Zero-shot, Seeds=42024.06 | 19 | 18 | 8 | 5 | |
| FBMode=Zero-shot, Seeds=42024.06 | 18 | 29 | 45 | 22 | |
| FBMode=Zero-shot, Seeds=42024.06 | 14 | 24 | 23 | 17 | |
| FDMMode=Zero-shot, Seeds=42024.06 | 12 | 28 | 21 | 34 |