支持向量机训练算法的实验比较

被引:5
作者
姬水旺
姬旺田
机构
[1] 陕西移动通信有限责任公司
关键词
统计学习理论; 支持向量机; 训练算法;
D O I
暂无
中图分类号
TP301.6 [算法理论];
学科分类号
摘要
SVM是基于统计学习理论的结构风险最小化原则的,它将最大分界面分类器思想和基于核的方法结合在一起,表现出了很好的泛化能力。并对目前的三种主流算法SVMlight,Bsvm与SvmFu在人脸检测、MNIST和USPS手写数字识别等应用中进行了系统比较。
引用
收藏
页码:18 / 20
页数:3
相关论文
共 32 条
[1]  
Everything Old is New Again: A Fresh Look at Historical Approaches in Machine Learning. Rifkin Ryan. . 2002
[2]  
Comparing Support Vector Machines with Gaussian Kernels to Radial Basis Function Classifiers. B Schlkopf,K K Sung,C J C Burges,et al. IEEE Transactions on Signal Processing . 1997
[3]  
Boosting the Margin: A New Explanation for the Effectiveness of Voting Methods. R Shapire,Y Freund,P Bartlett,et al. The Annals of Statistics . 1998
[4]  
ATrainingAlgorithmforOptimalMarginClassifiers. BEBoser,IMGuyon,VVapnik. . 1992
[5]  
AUnifiedFrameworkforRegularizationNetworksandSupportVectorMachines. TEvgeniou,MPontil,TPoggio. MassachusettsInstituteofTechnology:A.I.MemoNo.1654 . 1999
[6]  
An improved Training algorithm for Support Vector Machines. E Osuna,R Freund,F Girosi. . 1997
[7]  
New York. V. Vapnik,Statistical Learning Theory. Wiley Series in Solution Chemistry . 1998
[8]  
A Theory of Learnable. L G Valiant. Communications of the ACM . 1986
[9]  
An Introduction to Kernel-based Learning Algorithms. K R Muller,S Mika,G Ratsch,et al. IEEE Transactions on Neural Networks . 2001
[10]  
Machine Learning for Science: State of the Art and Future Prospects. E Mjolsmess,D DeCoste. Science . 2001