Offline Reinforcement Learning on Pointmass DeepMind Control Suite (test)
944.46Performance (Top Left)HILP
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| HILPSetting=Zero-shot offline RL2026.03 | 944.46 | 96.04 | 192.34 | 0.17 | |
| FBSetting=Zero-shot offline RL2026.03 | 897.83 | 274.95 | 517.23 | 19.37 | |
| RLDPSetting=Zero-shot offline RL2026.03 | 890.41 | 795.47 | 805.17 | 193.38 | |
| PSMSetting=Zero-shot offline RL2026.03 | 831.43 | 730.27 | 451.38 | 43.29 | |
| LaplaceSetting=Zero-shot offline RL2026.03 | 713.46 | 581.14 | 689.05 | 21.29 |