期刊
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
卷 31, 期 4, 页码 629-640出版社
IEEE COMPUTER SOC
DOI: 10.1109/TKDE.2018.2842023
关键词
Robust linear discriminant analysis; dimensionality reduction; L-2,L-1-norm minimization
类别
资金
- National Natural Science Foundation of China [61402002, 61502002, 61300057]
- Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry [48,2014-1685]
- US Natural Science Foundation of Anhui Province [1408085QF120, 1408085MKL94]
- Key Natural Science Project of Anhui Provincial Education Department [KJ2016A040]
- Open Project of IAT Collaborative Innovation Center of Anhui University [ADXXBZ201511]
Dimensionality reduction is a critical technology in the domain of pattern recognition, and linear discriminant analysis (LDA) is one of the most popular supervised dimensionality reduction methods. However, whenever its distance criterion of objective function uses L-2-norm, it is sensitive to outliers. In this paper, we propose a new formulation of linear discriminant analysis via joint L-2,L-1-norm minimization on objective function to induce robustness, so as to efficiently alleviate the influence of outliers and improve the robustness of proposed method. An efficient iterative algorithm is proposed to solve the optimization problem and proved to be convergent. Extensive experiments are performed on an artificial data set, on UCI data sets, and on four face data sets, which sufficiently demonstrates the efficiency of comparing to other methods and robustness to outliers of our approach.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据