Does constraining the search space of GA always help?: the case of balanced crossover operators

Abstract

In this paper, we undertake an investigation on the effect of balanced and unbalanced crossover operators against the problem of finding non-linear balanced Boolean functions: we consider three different balanced crossover operators and compare their performances with classic one-point crossover. The statistical comparison shows that the use of balanced crossover operators gives GA a definite advantage over one-point crossover.

Publication
Genetic and Evolutionary Computation Conference, GECCO 2019, Prague, Czech Republic, July 13-17, 2019, Companion Material Proceedings
Date

Note: this paper has been extended in this preprint.