A New Opposition-based Differential Evolution Variant Based on the Best and Worst Individuals

Abstract

Considering the drawbacks of the traditional Opposition-based Differential Evolution (ODE), a new variant of ODE is presented. It makes full use of the information in the best and worst individuals in the current population to learn the opposite estimates, and follows the same framework of the traditional ODE in population initialization and generation jumping. The proposed approach is validated using the benchmark functions chosen from CEC 2015. Empirical results show that the proposed approach gains a better performance when compare to classical DE and ODE.