Shade mutation strategy analysis via dynamic simulation in complex network

Abstract

This paper presents a novel approach to visualizing Evolutionary Algorithm (EA) dynamic in complex network and analyses the greediness of "current-topbest/1" mutation strategy used in state-of-art Differential Evolution (DE) algorithm-Success-History based Adaptive DE (SHADE) on CEC2015 benchmark set of test functions. Provided analysis suggests that the greediness might not be the optimal approach for guiding the evolution. © ECMS Zita Zoltay Paprika, Péter Horák, Kata Váradi,Péter Tamás Zwierczyk, Ágnes Vidovics-Dancs, János Péter Rádics (Editors).