Reinforcement Learning on Halfcheetah medium
5,168ReturnGELATO
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GELATOvariant=metric2021.02 | 5,168 | — | |
| MOPO2021.02 | 4,974 | — | |
| Imitation2021.02 | 4,201 | — | |
| GELATOvariant=L2021.02 | 4,096 | — | |
| Data Score2021.02 | 3,945 | — | |
| MBPO2021.02 | 3,228 | — | |
| SAC2021.02 | -839 | — | |
| Fixed-0.0Mixing Ratio=0.0, Backbone=Proto2026.05 | — | 90.8 | |
| Fixed-0.1Mixing Ratio=0.1, Backbone=Proto2026.05 | — | 92.55 | |
| Fixed-0.2Mixing Ratio=0.2, Backbone=Proto2026.05 | — | 94.6 | |
| Fixed-0.3Mixing Ratio=0.3, Backbone=Proto2026.05 | — | 92.06 | |
| Fixed-0.4Mixing Ratio=0.4, Backbone=Proto2026.05 | — | 91.41 | |
| Fixed-0.5Mixing Ratio=0.5, Backbone=Proto2026.05 | — | 89.89 | |
| ROADBackbone=Proto2026.05 | — | 94.88 |