Continuous probabilistic model building genetic network programming using reinforcement learning

Xianneng Li, Kotaro Hirasawa

Abstract: Highlights•This paper proposes a novel continuous estimation of distribution algorithm (EDA).•A recent EDA named PMBGNP is extended from discrete domain to continuous domain.•Reinforcement Learning (RL) is applied to construct the probabilistic model.•Experiments on real mobile robot control show the superiority of the proposed algorithm.•It bridges the gap between EDA and RL.