Differential Evolution with adaptive population size

Differential Evolution (DE) is one of the evolutionary algorithms under active research. It has been successfully applied to many real-world problems. The performance of DE highly depends on the population size Np. An improper selection of Np may result in premature convergence or waste of computational resources. In this paper, we proposed a novel method to adaptively control the population size of DE. With this method users do not need to set the Np parameter for DE. The proposed algorithm DEAPS is compared with the conventional DE with different population sizes. DEAPS demonstrates encouraging results on its capability of adaption for seven problems of benchmark test functions.