Offline Reinforcement Learning on D4RL Adroit (12 Tasks)
59Success RateReBRAC
Evaluation results
| Method | Links | |
|---|---|---|
| ReBRACPolicy Type=Gaussian Policies2025.10 | 59 | |
| IQLPolicy Type=Gaussian Policies2025.10 | 53 | |
| IFQLPolicy Type=Flow Policies2025.10 | 52 | |
| FQLPolicy Type=Flow Policies2025.10 | 52 | |
| CODACPolicy Type=Flow Policies2025.10 | 52 | |
| FBRACPolicy Type=Flow Policies2025.10 | 50 | |
| IQNPolicy Type=Flow Policies2025.10 | 50 | |
| Value FlowsPolicy Type=Flow Policies2025.10 | 50 | |
| BCPolicy Type=Gaussian Policies2025.10 | 48 | |
| C51Policy Type=Flow Policies2025.10 | 48 |