Reinforcement Learning on Hopper-Bodyparts Start OpenAI Gym Mujoco (initialization)
181.94Average RewardT-HiP-MDP
Evaluation results
| Method | Links | |
|---|---|---|
| T-HiP-MDPsamples per iteration=1x2022.10 | 181.94 | |
| LPG-FTWsamples per iteration=2x2022.10 | 166 | |
| EWCsamples per iteration=2x2022.10 | 133.95 | |
| VBLRLsamples per iteration=1x2022.10 | 93.27 | |
| Single-task MBRLmode=train from scratch2022.10 | 19.12 |