计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (29): 41-42.DOI: 10.3778/j.issn.1002-8331.2009.29.012

• 研究、探讨 • 上一篇    下一篇

r-SVR的PSO求解方法

方益民1,徐保国2   

  1. 1.江南大学 信息学院,江苏 无锡 214063
    2.江南大学 信控学院,江苏 无锡 214063
  • 收稿日期:2008-05-28 修回日期:2008-07-29 出版日期:2009-10-11 发布日期:2009-10-11
  • 通讯作者: 方益民

PSO method for r-SVR

FANG Yi-min1,XU Bao-guo2   

  1. 1.The School of Information Technology at Jiangnan University,Wuxi,Jiangsu 214063,China
    2.The School of Communications and Control Engineering at Jiangnan University,Wuxi,Jiangsu 214063,China
  • Received:2008-05-28 Revised:2008-07-29 Online:2009-10-11 Published:2009-10-11
  • Contact: FANG Yi-min

摘要: 针对r接近1时牛顿法不可求解r-SVR的问题,提出了r-SVR的PSO求解方法,推导出了PSO求解r-SVR的适应函数,给出了PSO求解r-SVR的算法和实验结果。结果表明,当r接近1时,牛顿法求解r-SVR失效,而PSO求解r-SVR的方法是有效的。

关键词: 支持向量回归机, r范数损失函数, 粒子群优化算法

Abstract: When r in r-SVR is less than 1,previous Newton descent method cannot be used for r-SVR.Hence,PSO algorithm is proposed to be used to solve r-SVR.In this paper,the fitness function is used in PSO for r-SVR and the relevant PSO algorithm is derived.Experimental results confirm that this method is effective.

Key words: support vector regression, r-loss function, particle swarm optimization

中图分类号: