AN IMPROVED STOCHASTIC GRADIENT ALGORITHM FOR PRINCIPAL COMPONENT ANALYSIS AND SUBSPACE TRACKING

Authors
DEHAENE J MOONEN M VANDEWALLE J
Citation
J. Dehaene et al., AN IMPROVED STOCHASTIC GRADIENT ALGORITHM FOR PRINCIPAL COMPONENT ANALYSIS AND SUBSPACE TRACKING, IEEE transactions on signal processing, 45(10), 1997, pp. 2582-2586
Citations number
10
Categorie Soggetti
Engineering, Eletrical & Electronic
Journal title
IEEE transactions on signal processing → ACNP
ISSN journal
1053587X
Volume
45
Issue
10
Year of publication
1997
Pages
2582 - 2586
Database
ISI
SICI code
1053-587X(1997)45:10<2582:AISGAF>2.0.ZU;2-X
Abstract
We propose a new stochastic gradient algorithm for principal component analysis and subspace tracking, requiring O(nm) operations per update , where n is the number of input signals, and m is the signal subspace dimension. A parallel version with problem size independent throughpu t is obtained at the expense of O(n(2)) additional flops.