Locomotion on DMControl Hopper
0.81Average Normalized ReturnTF-CLAM
Evaluation results
| Method | Links | |
|---|---|---|
| TF-CLAMInput modality=State-based2025.05 | 0.81 | |
| BC-ExpertInput modality=State-based, Privileged access=labeled expert data2025.05 | 0.76 | |
| MLP-CLAMInput modality=State-based2025.05 | 0.64 | |
| VPTInput modality=State-based2025.05 | 0.41 | |
| BC-ALInput modality=State-based2025.05 | 0.35 | |
| LAPAInput modality=State-based2025.05 | 0.3 | |
| LAPOInput modality=State-based2025.05 | 0.24 | |
| DynaMoInput modality=State-based2025.05 | 0.22 |