There is ALife beyond convergence: using a dual sharing to adapt intime dependent optimization

摘要

Using a time dependent optimization problem as a benchmark wepropose to improve the genetic algorithm's adaptiveness by anappropriate diversity management dynamics. The so-called Dual Sharing GAis an extension of a former algorithm based on the use of a genotype tophenotype mapping. We experimentally evaluate the benefits over similarapproaches to diversity management. We also analyze more precisely thedynamics exhibited at both phenotypic and genotypic levels in comparisonwith our previous approach