Physical state prediction on DeepMind Control Suite Go2 Easy tasks (random policy)
0.041MSEMoSim
Evaluation results
| Method | Links | |
|---|---|---|
| MoSimHorizon=16, Training data source=random policy data, Training approach=multistage approach2025.04 | 0.041 | |
| MoSimHorizon=16, Training data source=random policy data2025.04 | 0.043 | |
| MoSimHorizon=100, Training data source=random policy data2025.04 | 0.1282 | |
| MoSimHorizon=100, Training data source=random policy data, Training approach=multistage approach2025.04 | 0.1401 | |
| DreamerV3Horizon=16, Prediction steps=5-step, Training data source=random policy data2025.04 | 0.3685 | |
| DreamerV3Horizon=100, Prediction steps=5-step, Training data source=random policy data2025.04 | 0.4165 | |
| DreamerV3Horizon=100, Prediction steps=1-step, Training data source=random policy data2025.04 | 0.9243 | |
| DreamerV3Horizon=16, Prediction steps=1-step, Training data source=random policy data2025.04 | 1.0097 |