Reinforcement Learning on Walker URLB (downstream)
729Flip Success ScoreAPT
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| APT2025.06 | 729 | 542 | 949 | 892 | 3,112 | |
| AMPED2025.06 | 686 | 517 | 947 | 886 | 3,036 | |
| AMPED2025.06 | 674 | 467 | 951 | 929 | 3,021 | |
| CIC2025.06 | 637 | 454 | 939 | 874 | 2,904 | |
| BeCL2025.06 | 625 | 435 | 953 | 818 | 2,831 | |
| CeSD2025.06 | 623 | 377 | 915 | 805 | 2,720 | |
| DDPG2025.06 | 531 | 327 | 905 | 736 | 2,499 | |
| ComSD2025.06 | 488 | 341 | 937 | 826 | 2,592 | |
| RND2025.06 | 483 | 371 | 892 | 792 | 2,538 | |
| Continuous Metraskill_variable=Gaussian distribution2025.06 | 373 | 274 | 781 | 502 | 1,930 | |
| Discrete Metraskill_variable=categorical2025.06 | 329 | 174 | 764 | 406 | 1,673 | |
| DIAYN2025.06 | 329 | 183 | 716 | 434 | 1,662 |