作者: W. Bradley Knox , A. Ross Otto , Peter Stone , Bradley C. Love
关键词: Partially observable Markov decision process 、 Social psychology 、 Optimal decision 、 Reinforcement learning 、 Psychology 、 Poison control 、 Ideal (set theory) 、 Actor model 、 Task (project management) 、 Stochastic game
摘要: In non-stationary environments, there is a conflict between exploiting currently favored options and gaining information by exploring lesser-known options that in the past have …