ALSHADE-jSO Algorithm Solving the GNBG-II Test Suite

This paper presents an extended abstract describing an entry into the GNBG-II benchmarking competition. We propose a hybrid variant of the differential evolution (DE) algorithm based on the added properties of ALSHADE and jSO, named ALSHADE-jSO. This algorithm uses the current-to-Amean/1 mutation strategy inspired by ALSHADE and the weighted mean strategy inspired by jSO to generate new solutions. A selection strategy inspired by ALSHADE, SHADE-based parameter adjustments, and linear population size reduction (LPSR) mechanism of LSHADE are added for better exploration and exploitation operations. The results show that for 9 out of 24 problem instances, ALSHADE-jSO gives a success rate of 100% and is capable of achieving a global solution for 15 problems.