Machine Learning Algorithm for the Fitness Landscape Learning Evolutionary Computation

Abstract

One of the most important issues for evolutionary computation (EC) is to consider fitness landscape and the number of fitness evaluations. Especially, reducing the number of fitness evaluations is required in applications of EC to various kind of problems. In this paper, we proposed a novel EC framework called the Fitness Landscape Learning Evolutionary Computation: FLLEC with surrogate model which can predict the ranks of two individuals using SVM. The effectiveness of the proposed method is confirmed by computer simulation taking an Nk-landscape problem and a knapsack problem as examples by Air GA which is one of the FLLEC.