multimodal action distributions

3 papers

• Flow-Based Policy for Online Reinforcement Learning
• FreqPolicy: Efficient Flow-based Visuomotor Policy via Frequency Consistency
• Towards Robust Zero-Shot Reinforcement Learning