Research on Bayesian Optimization Algorithm Selection Strategy

摘要：

Probability model accuracy is the base of Bayesian Optimization Algorithm and data sample is the base of construction accuracy model. So sample strategy is critical for the algorithm. In test, tournament selection,truncation selection and proportional selection are adapted to deal with typical dependency-free function, bivariate dependencies function and multivariate dependencies function. The result shows that tournament selection is the best selection strategy for Bayesian Optimization Algorithm, truncation selection and proportional selection are unsuitable for the algorithm.

关键词： Bayesian Optimization Algorithm tournament selection truncation selection proportional selection

作者: Jiang min Chen Yimin

作者单位: School of mechanical and Automation Engineering Shanghai Institute of Technology No.120 Caobao Road, School of computer Engineering & Science Shanghai University No.149 Yanchang Road,Shanghai,China

会议类型: 国际会议

会议名称: 2010 IEEE信息与自动化国际会议(ICIA 2010)

会议地点: 哈尔滨

会议语种:英文

页码: 1-4

在线出版日期: 2010-06-20（万方平台首次上网日期，不代表论文的发表时间）