Reinforcement Learning on Hopper-Bodyparts OpenAI Gym Mujoco (train)
296.34Average RewardVBLRL
Evaluation results
| Method | Links | |
|---|---|---|
| VBLRLsamples per iteration=1x2022.10 | 296.34 | |
| LPG-FTWsamples per iteration=2x2022.10 | 256.13 | |
| T-HiP-MDPsamples per iteration=1x2022.10 | 227.5 | |
| EWCsamples per iteration=2x2022.10 | 139.01 | |
| Single-task MBRLmode=train from scratch2022.10 | 21.25 |