2018/02/28 by Yongsheng Liang, Liang, Yongsheng, Zhigang Ren +5
Computer Science · #Advanced Multi-Objective Optimization Algorithms #Evolutionary Algorithms and Applications #FOS: Computer and information sciences #Metaheuristic Optimization Algorithms Research #Neural and Evolutionary Computing (cs.NE)
paper · pdf · doi:10.48550/arxiv.1803.00986
openalex publication_date 2018/02/28 · openalex created_date 2022/02/25 · openalex updated_date 2026/07/28
As a model-based evolutionary algorithm, estimation of distribution algorithm\n(EDA) possesses unique characteristics and has been widely applied to global\noptimization. However, traditional Gaussian EDA (GEDA) may suffer from\npremature convergence and has a high risk of falling into local optimum when\ndealing with multimodal problem. In this paper, we first attempts to improve\nthe performance of GEDA by utilizing historical solutions and develops a novel\narchive-based EDA variant. The use of historical solutions not only enhances\nthe search efficiency of EDA to a large extent, but also significantly reduces\nthe population size so that a faster convergence could be achieved. Then, the\narchive-based EDA is further integrated with a novel adaptive clustering\nstrategy for solving multimodal optimization problems. Taking the advantage of\nthe clustering strategy in locating different promising areas and the powerful\nexploitation ability of the archive-based EDA, the resultant algorithm is\nendowed with strong capability in finding multiple optima. To verify the\nefficiency of the proposed algorithm, we tested it on a set of well-known\nniching benchmark problems and compared it with several state-of-the-art\nniching algorithms. The experimental results indicate that the proposed\nalgorithm is competitive.\n