Sequential minimal optimization for SVM with pinball loss

Sequential minimal optimization for SVM with pinball loss
复制标题

具有 pinball 损失的 SVM 的顺序最小优化

DOI:
10.1016/j.neucom.2014.08.033
复制
发表时间:
2015-02
期刊:
影响因子:
6
通讯作者:
Johan A.K. Suykens
Johan A.K. Suykens
中科院分区:
计算机科学2区
文献类型:
--
作者:
Xiaolin Huang;Lei Shi;Johan A.K. Suykens

文献摘要

参考文献

被引文献

相似文献

为了追求对特征噪声的不敏感性和对重采样的稳定性,将经典支持向量机中的铰链损失替换为弹球损失,建立了一种新的支持向量机,称为apin-SVM。虽然使用了不同的损失函数,但pin-SVM具有与经典SVM相似的结构。具体地说,pin-SVM的对偶问题是一个带有框约束的二次规划问题,序列最小优化(SMO)技术是适用的。在本文中,我们建立SMO算法的pin-SVM及其稀疏版本。在真实数据集上的数值实验表明了pin-SVM的良好性能和SMO方法的有效性。
To pursue the insensitivity to feature noise and the stability to re-sampling, a new type of support vector machine (SVM) has been established via replacing the hinge loss in the classical SVM by the pinball loss and was hence called apin-SVM. Though a different loss function is used, pin-SVM has a similar structure as the classical SVM. Specifically, the dual problem of pin-SVM is a quadratic programming problem with box constraints, for which the sequential minimal optimization (SMO) technique is applicable. In this paper, we establish SMO algorithms for pin-SVM and its sparse version. The numerical experiments on real-life data sets illustrate both the good performance of pin-SVMs and the effectiveness of the established SMO methods.
DOI: 10.1198/jasa.2006.s143
发表时间: 2006-12
影响因子: 3.7
作者:
J. Jurečková
通讯作者: J. Jurečková
DOI: 10.1023/a:1012431217818
发表时间: 2002-01-01
期刊: MACHINE LEARNING
影响因子: 7.5
作者:
Keerthi, SS;Gilbert, EG
通讯作者: Gilbert, EG
DOI: 10.3150/10-bej267
发表时间: 2011-02-01
期刊: BERNOULLI
影响因子: 1.5
作者:
Steinwart, Ingo;Christmann, Andreas
通讯作者: Christmann, Andreas
DOI: --
发表时间: 2006-12
期刊: J. Mach. Learn. Res.
影响因子: --
作者:
D. Hush;P. Kelly;C. Scovel;Ingo Steinwart
通讯作者: D. Hush;P. Kelly;C. Scovel;Ingo Steinwart
DOI: --
发表时间: 2005-12
期刊: J. Mach. Learn. Res.
影响因子: --
作者:
Rong-En Fan;Pai-Hsuen Chen;Chih-Jen Lin
通讯作者: Rong-En Fan;Pai-Hsuen Chen;Chih-Jen Lin