Task Generalization on Meta-World ML-10 (train)
97.8Success RateMZ+Recon
Evaluation results
| Method | Links | |
|---|---|---|
| MZ+ReconInput source=pixels, Environment steps=50M2021.11 | 97.8 | |
| MuZeroInput source=pixels, Environment steps=50M2021.11 | 97.6 | |
| RL2Input source=states, Environment steps=400M2021.11 | 86.9 | |
| Q-LearningInput source=pixels, Environment steps=50M2021.11 | 85.2 | |
| MAMLInput source=states, Environment steps=400M2021.11 | 44.4 |