Meta Generative Flow Networks with personalization for task-specific adaptation

Xinyuan Ji, Xu Zhang, Wei Xi, Haozhi Wang, Olga Gadyatskaya, Yinchuan Li

Abstract: Highlights•A method is proposed to boost the exploration performance of reinforcement learning on new tasks.•A personalized method is proposed to boost reinforcement learning on distinct tasks.•The convergence analysis is established for the personalized method.