Offline Reinforcement Learning on Rotational Reacher (Mixed)
0.96ReturnDAC-Output
Evaluation results
| Method | Links | |
|---|---|---|
| DAC-OutputAlgorithm=IQL, Zero-shot Policy Transfer=true, Rotation Group=C4 (90°)2026.07 | 0.96 | |
| Aug-D-OnlineAlgorithm=IQL, Zero-shot Policy Transfer=true, Rotation Group=C4 (90°)2026.07 | 0.71 | |
| Aug-DAlgorithm=IQL, Zero-shot Policy Transfer=true, Rotation Group=C4 (90°)2026.07 | 0.68 | |
| DAC-LatentAlgorithm=IQL, Zero-shot Policy Transfer=true, Rotation Group=C4 (90°)2026.07 | 0.67 | |
| No DAAlgorithm=IQL, Zero-shot Policy Transfer=true, Data Augmentation=None2026.07 | 0.34 |