A study on the effect of multi-parent recombination in real coded genetic algorithms

Abstract

We investigate real coded genetic algorithms in which more than two parents are involved in recombination operation. We propose three types of multi-parent recombination operators; the center of mass crossover (CMX), multi-parent feature-wise crossover (MFX), and seed crossover (SX). Each of these operators is a natural generalization of 2-parent recombination operator. These operators are evaluated on several test functions. The results showed clearly that multi-parent recombinations lead to better performance, although the performance improvement for different techniques were found to be dependent on problems.