Racial Harmony and Function Optimization in Genetic Algorithms—The Races Genetic Algorithm
Conor Ryan · The MIT Press eBooks · 1995
This paper introduces a modified genetic algorithm, RGA, which, through the use of races, maintains a very diverse population. RGA is shown to be particularly good at solving mulitmodal functions and outperforms current methods, both in discovering solutions and maintaining them. 1 INTRODUCTION Many papers investigating genetic algorithms concern themselves with the location and optimization of a single global maximum (Beasley, D. et. al. 93). The solution landscape of these problems usually contains a single, dominant peak which the population converges on. But, many problems of interest contain several peaks in the fitness landscape. Sometimes, only one of these peaks will be of interest, but often (Spears 94) (Deb 89a) (Beasley, D. et. al. 93) it is desirable to find all peaks of the landscape. Unlike problems in which the peak of interest changes with time, (Ryan 94c) (Hillis 92) (Siegel 94), it is important to maintain solutions at all peaks in the landscape, ensuring that one p...