A Comparison of Crossover and Mutation in Genetic Programming

Abstract

In [Luke and Spector 1997] we presented a comprehensive suite of data comparing GP crossover and point mutation over four domains and a wide range of parameter settings. Unfortunately, the results were marred by statistical flaws. This revision of the study eliminates these flaws, with three times as much the data as the original experiments had. Our results again show that crossover does have some advantage over mutation given the right parameter settings (primarily larger population sizes), though the difference between the two surprisingly small. Further, the results are complex, suggesting that the big picture is more complicated than is commonly believed. 1