Reinforcement Learning on DeepMind Control Walker Run
205,000Steps to 75% ReturnMind Dreamer
Evaluation results
| Method | Links | |
|---|---|---|
| Mind DreamerHorizon=52026.05 | 205,000 | |
| Mind DreamerHorizon=152026.05 | 244,300 | |
| Dreamer V32026.05 | 335,000 | |
| Mind DreamerHorizon=102026.05 | 335,000 | |
| Dreamer V22026.05 | 377,500 |