标题:Gaussian Cauchy Differential Evolution for Global Optimization
作者:Zhang, Qingke; Zhang, Huaxiang; Yang, Bo; Hu, Yupeng
通讯作者:Zhang, QK;Zhang, HX;Zhang, QK;Zhang, HX;Zhang, Qingke
作者机构:[Zhang, Qingke; Zhang, Huaxiang] Shandong Normal Univ, Sch Informat Sci & Engn, Jinan, Shandong, Peoples R China.; [Zhang, Qingke; Zhang, Huaxiang] 更多
会议名称:1st CCF International Conference on Artificial Intelligence (CCF-ICAI)
会议日期:AUG 09-10, 2018
来源:ARTIFICIAL INTELLIGENCE (ICAI 2018)
出版年:2018
卷:888
页码:166-182
DOI:10.1007/978-981-13-2122-1_13
关键词:Evolutionary computation; Differential evolution; Bare-bones particle; swarm optimization; Adaptive parameters; Global optimization
摘要:Differential evolution (DE) has been proven to be a powerful and efficient stochastic search technique for global numerical optimization. However, choosing the optimal control parameters of DE is a time-consuming task because they are problem depended. DE may have a strong ability in exploring the search space and locating the promising area of global optimum but may be slow at exploitation. Thus, in this paper, we propose a Gaussian Cauchy differential evolution (GCDE). It is a hybrid of a modified bare-bones swarm optimizers and the differential evolution algorithm. It takes advantage of the good exploration searching ability of DE and the good exploitation ability of bare-bones optimization. Moreover, the parameters in GCDE are generated by the function of Gaussian distribution and Cauchy distribution. In addition, the parameters dynamically change according to the quality of the current search solution. The performance of proposed method is compared with three differential evolution algorithms and three bare-bones technique based optimizers. Comprehensive experimental results show that the proposed approach is better than, or at least comparable to, other classic DE variants when considering the quality of search solutions on a set of benchmark problems.
收录类别:CPCI-S;EI;SCOPUS
资源类型:会议论文;期刊论文
原文链接:https://www.scopus.com/inward/record.uri?eid=2-s2.0-85052056236&doi=10.1007%2f978-981-13-2122-1_13&partnerID=40&md5=a7fb05542a0015249c318a1d0a69d0b4
TOP