Robust L1-norm two-dimensional linear discriminant analysis
Robust L1-norm two-dimensional linear discriminant analysis
复制标题
鲁棒 L1 范数二维线性判别分析
DOI：
10.1016/j.neunet.2015.01.003
复制
发表时间：
2015-05-01
期刊：
影响因子：
7.8
通讯作者：
Deng, Nai-Yang
中科院分区：
文献类型：
--
作者：
Li, Chun-Na;Shao, Yuan-Hai;Deng, Nai-Yang
本文提出了一种具有鲁棒性能的L1范数二维线性鉴别分析（L1-2DLDA）。与传统的二维L2范数线性判别分析（L2-2DLDA）将优化问题转化为广义特征值问题不同，L1-2DLDA中的优化问题通过一个简单的合理迭代技术求解，并保证了其收敛性.与L2-2DLDA相比，由于使用了L1范数，因此L1-2DLDA对离群点和噪声具有更好的鲁棒性。我们在玩具样本和人脸数据集上的初步实验支持了这一点，这些实验表明我们的L1-2DLDA比L2-2DLDA有所改进。(C)2015爱思唯尔有限公司版权所有。
In this paper, we propose an L1-norm two-dimensional linear discriminant analysis (L1-2DLDA) with robust performance. Different from the conventional two-dimensional linear discriminant analysis with L2-norm (L2-2DLDA), where the optimization problem is transferred to a generalized eigenvalue problem, the optimization problem in our L1-2DLDA is solved by a simple justifiable iterative technique, and its convergence is guaranteed. Compared with L2-2DLDA, our L1-2DLDA is more robust to outliers and noises since the L1-norm is used. This is supported by our preliminary experiments on toy example and face datasets, which show the improvement of our L1-2DLDA over L2-2DLDA. (C) 2015 Elsevier Ltd. All rights reserved.