General Subspace Learning With Corrupted Training Data Via Graph Embedding

	General Subspace Learning With Corrupted Training Data Via Graph Embedding
	Bao, Bing-Kun1 ; Liu, Guangcan 2; Hong, Richang 3; Yan, Shuicheng 4; Xu, Changsheng1
发表期刊	IEEE TRANSACTIONS ON IMAGE PROCESSING
	2013-11-01
卷号	22 期号:11 页码:4380-4393
文章类型	Article
摘要	We address the following subspace learning problem: supposing we are given a set of labeled, corrupted training data points, how to learn the underlying subspace, which contains three components: an intrinsic subspace that captures certain desired properties of a data set, a penalty subspace that fits the undesired properties of the data, and an error container that models the gross corruptions possibly existing in the data. Given a set of data points, these three components can be learned by solving a nuclear norm regularized optimization problem, which is convex and can be efficiently solved in polynomial time. Using the method as a tool, we propose a new discriminant analysis (i.e., supervised subspace learning) algorithm called Corruptions Tolerant Discriminant Analysis (CTDA), in which the intrinsic subspace is used to capture the features with high within-class similarity, the penalty subspace takes the role of modeling the undesired features with high between-class similarity, and the error container takes charge of fitting the possible corruptions in the data. We show that CTDA can well handle the gross corruptions possibly existing in the training data, whereas previous linear discriminant analysis algorithms arguably fail in such a setting. Extensive experiments conducted on two benchmark human face data sets and one object recognition data set show that CTDA outperforms the related algorithms.
关键词	Subspace Learning Corrupted Training Data Discriminant Analysis Graph Embedding
WOS标题词	Science & Technology ; Technology
关键词[WOS]	NONLINEAR DIMENSIONALITY REDUCTION ; ROBUST ; FRAMEWORK ; PURSUIT ; SYSTEMS
收录类别	SCI
语种	英语
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:000324597800018
引用统计	被引频次：25[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/2840
专题	多模态人工智能系统全国重点实验室_多媒体计算
作者单位	1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 2.Univ Illinois, Champaign, IL 61820 USA 3.Hefei Univ Technol, Sch Comp & Informat, Hefei 230009, Peoples R China 4.Natl Univ Singapore, Dept Elect & Comp Engn, Singapore 10000, Singapore
第一作者单位	模式识别国家重点实验室
推荐引用方式 GB/T 7714	Bao, Bing-Kun,Liu, Guangcan,Hong, Richang,et al. General Subspace Learning With Corrupted Training Data Via Graph Embedding[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2013,22(11):4380-4393.
APA	Bao, Bing-Kun,Liu, Guangcan,Hong, Richang,Yan, Shuicheng,&Xu, Changsheng.(2013).General Subspace Learning With Corrupted Training Data Via Graph Embedding.IEEE TRANSACTIONS ON IMAGE PROCESSING,22(11),4380-4393.
MLA	Bao, Bing-Kun,et al."General Subspace Learning With Corrupted Training Data Via Graph Embedding".IEEE TRANSACTIONS ON IMAGE PROCESSING 22.11(2013):4380-4393.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
General Subspace Lea（3715KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览下载