Cooperative co-evolution with correlation identification grouping for large scale function optimization

Cooperative co-evolutionary (CC) architectures provide a framework for solving large scale function optimization problems, by decomposing variables into different groups as subproblems, solving the subproblems, and then reintegrating the solutions. But there is no systematic method for how to decomposing variables, which is a major abstacle for CC framework. This paper provides a correlation identification technique for variables grouping; and combining with Differential Evolution (DE), a Cooperative co-evolutionary differential evolution algorithm with correlation identification grouping (DECC-CIG) is presented. The performance of DECC-CIG is compared with DECC and DECC-NW to highlight its benefits.