Reinforcement Learning on Walker-Gravity OpenAI Gym Mujoco (train)
333.65Average RewardVBLRL
Evaluation results
| Method | Links | |
|---|---|---|
| VBLRLsamples per iteration=1x2022.10 | 333.65 | |
| T-HiP-MDPsamples per iteration=1x2022.10 | 217.94 | |
| Single-task MBRLmode=train from scratch2022.10 | 210.41 | |
| EWCsamples per iteration=2x2022.10 | 181.76 | |
| LPG-FTWsamples per iteration=2x2022.10 | 140.75 |