Reinforcement Learning on Walker-Bodyparts OpenAI Gym Mujoco (train)
311.41Average RewardVBLRL
Evaluation results
| Method | Links | |
|---|---|---|
| VBLRLsamples per iteration=1x2022.10 | 311.41 | |
| T-HiP-MDPsamples per iteration=1x2022.10 | 266.97 | |
| EWCsamples per iteration=2x2022.10 | 206.97 | |
| Single-task MBRLmode=train from scratch2022.10 | 196.42 | |
| LPG-FTWsamples per iteration=2x2022.10 | 165.75 |