A Hierarchical Cooperative Evolutionary Algorithm

摘要

To successfully search multiple coadaptive subcomponents in a solution, we developed a novel cooperative evolutionary algorithm based on a new computational multilevel selection framework. This algorithm constructs cooperative solutions hierarchically by implementing the idea of group selection. We show that this simple and straightforward algorithm is able to accelerate evolutionary speed and improve solution accuracy on string covering problems as compared to other EAs used in literature. In addition, the structure of the solution and the roles played by each subcomponent in the solution emerge as a result of evolution without human interference.