MRPO: Magnitude-Regularized Policy Optimization via L1 Constraints

ICML

Wei, Han and Yuanxing, Liu and Mingda, Li and Ruiyu, Xiao and Weinan, Zhang and Ting, Liu