EVD Dualdating Based Online Subspace Learning
Joint Authors
Jin, Bo
Jing, Zhongliang
Zhao, Haitao
Source
Mathematical Problems in Engineering
Issue
Vol. 2014, Issue 2014 (31 Dec. 2014), pp.1-21, 21 p.
Publisher
Hindawi Publishing Corporation
Publication Date
2014-07-24
Country of Publication
Egypt
No. of Pages
21
Main Subjects
Abstract EN
Conventional incremental PCA methods usually only discuss the situation of adding samples.
In this paper, we consider two different cases: deleting samples and simultaneously adding and deleting samples.
To avoid the NP-hard problem of downdating SVD without right singular vectors and specific position information, we choose to use EVD instead of SVD, which is used by most IPCA methods.
First, we propose an EVD updating and downdating algorithm, called EVD dualdating, which permits simultaneous arbitrary adding and deleting operation, via transforming the EVD of the covariance matrix into a SVD updating problem plus an EVD of a small autocorrelation matrix.
A comprehensive analysis is delivered to express the essence, expansibility, and computation complexity of EVD dualdating.
A mathematical theorem proves that if the whole data matrix satisfies the low-rank-plus-shift structure, EVD dualdating is an optimal rank-k estimator under the sequential environment.
A selection method based on eigenvalues is presented to determine the optimal rank k of the subspace.
Then, we propose three incremental/decremental PCA methods: EVDD-IPCA, EVDD-DPCA, and EVDD-IDPCA, which are adaptive to the varying mean.
Finally, plenty of comparative experiments demonstrate that EVDD-based methods outperform conventional incremental/decremental PCA methods in both efficiency and accuracy.
American Psychological Association (APA)
Jin, Bo& Jing, Zhongliang& Zhao, Haitao. 2014. EVD Dualdating Based Online Subspace Learning. Mathematical Problems in Engineering،Vol. 2014, no. 2014, pp.1-21.
https://search.emarefa.net/detail/BIM-471585
Modern Language Association (MLA)
Jin, Bo…[et al.]. EVD Dualdating Based Online Subspace Learning. Mathematical Problems in Engineering No. 2014 (2014), pp.1-21.
https://search.emarefa.net/detail/BIM-471585
American Medical Association (AMA)
Jin, Bo& Jing, Zhongliang& Zhao, Haitao. EVD Dualdating Based Online Subspace Learning. Mathematical Problems in Engineering. 2014. Vol. 2014, no. 2014, pp.1-21.
https://search.emarefa.net/detail/BIM-471585
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references
Record ID
BIM-471585