Enhanced Differential Evolution Entirely Parallel Method for Biomedical Applications

A considerable enhancement is proposed for the Differential Evolution Entirely Parallel DEEP method developed recently. A new selection rule was implemented in order to increase the robustness of DEEP. To simplify the approach a population is not divided now into branches, instead of it, several oldest individuals are substituted with the same number of the best ones after the predefined number of iterations. The individuals are selected on the basis of the number of generations, in which they survived without any change. We demonstrate how the enhanced DEEP provides new solutions to problems with several objective functions.