ABSTRACT

This chapter focuses on the class of algorithms called cellular evolutionary algorithms (cEAs). Here, we present the canonical algorithm and suggest the interesting variants targeted to solve complex problems accurately with a minimum effort for customization. These techniques, also called diffusion or finegrained models, have been popularized, among others, by the early work of Gorges-Schleuter [1] and Manderick and Spiessens [2]. The basic idea is to add some structure to the population of tentative solutions. The pursued effect is to improve on the diversity and exploration capabilities of the algorithm while, still admitting an easy combination with local search and other search techniques to improve on exploitation.