Large-scale Evolutionary Multi-objective Optimization Assisted by Directed Sampling

Abstract

It is particularly challenging for evolutionary algorithms to quickly converge to the Pareto front in large-scale multi-objective optimization. To tackle this problem, this paper proposes a large-scale multi-objective evolutionary algorithm assisted by some selected individuals generated by directed sampling. At each generation, a set of individuals closer to the ideal point are chosen for performing a directed sampling in the decision space, and those non-dominated ones of the sampled solutions are used to assist the reproduction to improve the convergence in evolutionary large-scale multi-objective optimization. In addition, elitist non-dominated sorting is adopted complementarily for environmental selection with a reference vector based method in order to maintain diversity of the population. Our experimental results show that the proposed algorithm is highly competitive on large-scale multi-objective optimization test problems with up to 5000 decision variables compared to five state-of-the-art multi-objective evolutionary algorithms.