An Evolutionary Multi- and Many-Objective Optimization Algorithm Based on ISDE + and Region Decomposition
Zixian Lin, Hai‐Lin Liu, Fangqing Gu · 2018
In this paper, we propose an evolutionary multi-and many-objective optimization algorithm combining ISDE+and region decomposition. It decomposes the objective space into a number of sub-regions by a set of direction vectors and independently calculates the indicator ISDE+by using the corresponding direction vector in each subregion. Thus, the convergence direction of each sub-region is relatively adjusted. In this way, the proposed algorithm can adapt to various of Pareto Front shapes. The inferior individuals are eliminated according to the value of ISDE+of each individual one by one. In the experiments, we compare the proposed algorithm with four evolutionary multi-and many-objective optimization algorithms on WFG series with different number of objectives. The result shows that the proposed algorithm promotes diversity and convergence.