An Adaptive Niching-Based Evolutionary Algorithm for Optimizing Multi-Modal Function

This paper presents a niching-based evolutionary algorithm for optimizing multi-modal optimization function. Provided that the potential optima are characterized by a relatively smaller objective value than their neighbors and by a relatively large distance from points with smaller objective values, we identify potential optima from individuals. Using them as seeds, a population is decomposed into a number of subpopulations without introducing new parameters. Moreover, we present an adaptive allocating strategy of assigning different computational resources to different subpopulations upon the fact that discovering different optima may have different computational difficulty. The proposed method is compared with three state-of-the-art multi-modal optimization approaches on a benchmark function set. The extensive experimental results demonstrate its efficacy.