Robust L1-norm non-parallel proximal support vector machine

Robust L1-norm non-parallel proximal support vector machine
复制标题

鲁棒L1范数非并行近端支持向量机

DOI:
10.1080/02331934.2014.994627
复制
发表时间:
2016-01
期刊:
影响因子:
2.2
通讯作者:
Deng Nai-Yang
Deng Nai-Yang
中科院分区:
数学3区
文献类型:
--
作者:
Li Chun-Na;Shao Yuan-Hai;Deng Nai-Yang

文献摘要

参考文献

被引文献

相似文献

在本文中,我们提出了一个强大的L1范数非并行邻近支持向量机(L1-NPSVM),其目的是提供一个强大的性能相比,GEPSVM的二进制分类,特别是对离群值的问题。提出的L1-NPSVM有三个主要性质。首先,与传统的GEPSVM求解两个广义特征值问题不同,我们的L1-NPSVM通过使用简单的合理迭代技术求解了一对L1-范数最优问题。其次,通过引入L1-范数,L1-NPSVM在很大程度上比GEPSVM对离群值具有更强的鲁棒性。第三,与GEPSVM相比,我们的L1-NPSVM不需要正则化参数。通过对一个简单的人工样本和UCI数据集的测试,证明了该方法的有效性,表明了GEPSVM的改进。
In this paper, we propose a robust L1-norm non-parallel proximal support vector machine (L1-NPSVM), which aims at giving a robust performance for binary classification in contrast to GEPSVM, especially for the problem with outliers. There are three mainly properties of the proposed L1-NPSVM. Firstly, different from the traditional GEPSVM which solves two generalized eigenvalue problems, our L1-NPSVM solves a pair of L1-norm optimal problems by using a simple justifiable iterative technique. Secondly, by introducing the L1-norm, our L1-NPSVM is more robust to outliers than GEPSVM to a great extent. Thirdly, compared with GEPSVM, no parameters need to be regularized in our L1-NPSVM. The effectiveness of the proposed method is demonstrated by tests on a simple artificial example as well as on some UCI datasets, which shows the improvements of GEPSVM.
基于L1范数最大化的线性判别分析
DOI: 10.1109/tip.2013.2253476
发表时间: 2013-03
期刊: IEEE Trans. On Image Processing
影响因子: --
作者:
Fujin Zhong;Jiashu Zhang;Defang Li
通讯作者: Defang Li
DOI: 10.1016/j.neunet.2012.09.004
发表时间: 2012-12
期刊: Neural networks : the official journal of the International Neural Network Society
影响因子: --
作者:
Zhiquan Qi;Ying-jie Tian;Yong Shi
通讯作者: Zhiquan Qi;Ying-jie Tian;Yong Shi
DOI: 10.1007/11596448_85
发表时间: 2005-09
期刊: The Fifth International Conference on Computer and Information Technology (CIT'05)
影响因子: --
作者:
Zhiquan Qi;Ying-jie Tian;N. Deng
通讯作者: Zhiquan Qi;Ying-jie Tian;N. Deng
DOI: --
发表时间: 1996
期刊: --
影响因子: --
作者:
C. Merz
通讯作者: C. Merz
DOI: --
发表时间: 1996
期刊: --
影响因子: --
作者:
Sameer A. Nene;S. Nayar;H. Murase
通讯作者: Sameer A. Nene;S. Nayar;H. Murase