期刊文献+

r-SVR的PSO求解方法

PSO method for r-SVR
下载PDF
导出
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。 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.
出处 《计算机工程与应用》 CSCD 北大核心 2009年第29期41-42,59,共3页 Computer Engineering and Applications
基金 国家高技术研究发展计划(863)(No.2006AA10Z248)~~
关键词 支持向量回归机 r范数损失函数 粒子群优化算法 support vector regression r-loss function particle swarm optimization
  • 相关文献

参考文献5

二级参考文献17

  • 1Cristianini N,Shawe-Taylor J. An Introduction to Support Vector Machines[M]. Cambridge University Press, 2000
  • 2Vapnik V. Statistical Learning Theory[M]. New York: Wiley,1998
  • 3Kwok J T,Tsang I W. Linear dependency between ε and the input noise in ε-support vector regression[J]. IEEE Transaction on Neural Networks, 2003(5)
  • 4Smola A J, Murata N, Scholkopf B, Müller K-R. Asymptotically optimal choice of ε-loss for support vector machines[A]. In:Proc. of the Intl. Conf. on Artificial Neural Networks[C]. 1998
  • 5Smola A J,Scholkopf B. A tutotial on support vector regression. Royal Holloway College, 1998, NeuroCOLT2 Technical Report NC2-TR-1998-030
  • 6Law M H,Kwok J T. Bayesian support vector regression[A]. In:Proc. of the English Intl. Workshop on Artificial Intelligence and Statistics[C]. Florida:Key West, 2001. 239~244
  • 7Gao J B, Gunn S R, Ham's C J. A probabilistic framework for SVM regression and Error Bar Estimation[J]. Machine Learning, 2002,46:71~89
  • 8Cherkassky V, Ma Yunqian. Practical selection of SVM parame ters and noise estimation for SVM regression[J]. Neural Networks, in press, 2003
  • 9阎平凡 张长水.人工神经网络和模拟进化计算[M].北京:清华大学出版社,2001..
  • 10沈永欢.实用数学手册[M].北京:科学出版社,2002..

共引文献13

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部