ExORL
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Zero-shot Reinforcement Learning | ExORL RND (Quadruped environment) v1 (test) | Jump Success758 | 12 | |
| Zero-shot Reinforcement Learning | ExORL RND Walker environment v1 (test) | Flip644 | 12 | |
| Offline Reinforcement Learning | ExORL | Cheetah Run Score104.7 | 9 | |
| Visual Control | ExORL Jaco Zero-shot RND | Reach Top Left48 | 8 | |
| Visual Control | ExORL Cheetah Zero-shot RND | Walk Score805 | 8 | |
| Zero-shot Reinforcement Learning | ExORL APS (Jaco environment) v1 (test) | Reach Bottom Left88 | 8 | |
| Zero-shot Reinforcement Learning | ExORL APS Cheetah environment v1 (test) | Run Backward383 | 8 | |
| Unsupervised Reinforcement Learning | ExORL jaco (4 tasks) zero-shot | Average Return23 | 6 | |
| Unsupervised Reinforcement Learning | ExORL quadruped zero-shot | Average Return645 | 6 | |
| Unsupervised Reinforcement Learning | ExORL cheetah (4 tasks) zero-shot | Average Return378 | 6 | |
| Unsupervised Reinforcement Learning | ExORL walker (4 tasks) zero-shot | Average Return619 | 6 | |
| Zero-shot Reinforcement Learning | ExORL APS Quadruped environment v1 (test) | Jump Score757 | 4 | |
| Zero-shot Reinforcement Learning | ExORL APS Walker environment v1 (test) | Flip Count573 | 4 |
Showing 13 of 13 rows