On Cooperative Coevolution and Global Crossover

TL;DR

This paper investigates the use of a global crossover operator in cooperative coevolutionary algorithms, showing potential improvements over traditional methods especially on rugged fitness landscapes.

Contribution

It introduces a perspective of viewing CCEAs without subproblem partnering as a form of global crossover, and explores its effects on rugged landscapes using the NK model.

Findings

Global crossover can improve CCEA performance on rugged landscapes

Removing subproblem partnering simplifies the algorithm structure

Results demonstrate advantages over traditional CCEAs on test functions

Abstract

Cooperative coevolutionary algorithms (CCEAs) divide a given problem in to a number of subproblems and use an evolutionary algorithm to solve each subproblem. This short paper is concerned with the scenario under which only a single, global fitness measure exists. By removing the typically used subproblem partnering mechanism, it is suggested that such CCEAs can be viewed as making use of a generalised version of the global crossover operator introduced in early Evolution Strategies. Using the well-known NK model of fitness landscapes, the effects of varying aspects of global crossover with respect to the ruggedness of the underlying fitness landscape are explored. Results suggest improvements over the most widely used form of CCEAs, something further demonstrated using other well-known test functions.