Hybrid cooperative co-evolution for the CEC15 benchmarks

In this paper, we test the performance of hybrid cooperative co-evolution (hCC) on the CEC15 benchmarks. In its initial stage, the method applies the recently introduced differential grouping to learn the problem variables´ inter-dependencies and separate the variables into groups of separable and non-separable ones. In its second stage, the method adopts different algorithms within the cooperative co-evolution (CC) framework to simultaneously optimize the generated groups. Results are reported for all required problem sizes.