COUPLED SINGULAR VALUE DECOMPOSITION OF A CROSS-COVARIANCE MATRIX

A. KAISER, W. Schenck, R. MÖLLER, International Journal of Neural Systems 20 (2012) 293–318.

Download
Es wurde kein Volltext hochgeladen. Nur Publikationsnachweis!
Artikel | Veröffentlicht | Englisch
Autor*in
KAISER, ALEXANDER; Schenck, WolframFH Bielefeld ; MÖLLER, RALF
Abstract
We derive coupled on-line learning rules for the singular value decomposition (SVD) of a cross-covariance matrix. In coupled SVD rules, the singular value is estimated alongside the singular vectors, and the effective learning rates for the singular vector rules are influenced by the singular value estimates. In addition, we use a first-order approximation of Gram-Schmidt orthonormalization as decorrelation method for the estimation of multiple singular vectors and singular values. Experiments on synthetic data show that coupled learning rules converge faster than Hebbian learning rules and that the first-order approximation of Gram-Schmidt orthonormalization produces more precise estimates and better orthonormality than the standard deflation method.
Erscheinungsjahr
Zeitschriftentitel
International Journal of Neural Systems
Band
20
Zeitschriftennummer
04
Seite
293-318
ISSN
eISSN
FH-PUB-ID

Zitieren

KAISER, ALEXANDER ; Schenck, Wolfram ; MÖLLER, RALF: COUPLED SINGULAR VALUE DECOMPOSITION OF A CROSS-COVARIANCE MATRIX. In: International Journal of Neural Systems Bd. 20, World Scientific Pub Co Pte Lt (2012), Nr. 04, S. 293–318
KAISER A, Schenck W, MÖLLER R. COUPLED SINGULAR VALUE DECOMPOSITION OF A CROSS-COVARIANCE MATRIX. International Journal of Neural Systems. 2012;20(04):293-318. doi:10.1142/S0129065710002437
KAISER, A., Schenck, W., & MÖLLER, R. (2012). COUPLED SINGULAR VALUE DECOMPOSITION OF A CROSS-COVARIANCE MATRIX. International Journal of Neural Systems, 20(04), 293–318. https://doi.org/10.1142/S0129065710002437
@article{KAISER_Schenck_MÖLLER_2012, title={COUPLED SINGULAR VALUE DECOMPOSITION OF A CROSS-COVARIANCE MATRIX}, volume={20}, DOI={10.1142/S0129065710002437}, number={04}, journal={International Journal of Neural Systems}, publisher={World Scientific Pub Co Pte Lt}, author={KAISER, ALEXANDER and Schenck, Wolfram and MÖLLER, RALF}, year={2012}, pages={293–318} }
KAISER, ALEXANDER, Wolfram Schenck, and RALF MÖLLER. “COUPLED SINGULAR VALUE DECOMPOSITION OF A CROSS-COVARIANCE MATRIX.” International Journal of Neural Systems 20, no. 04 (2012): 293–318. https://doi.org/10.1142/S0129065710002437.
A. KAISER, W. Schenck, and R. MÖLLER, “COUPLED SINGULAR VALUE DECOMPOSITION OF A CROSS-COVARIANCE MATRIX,” International Journal of Neural Systems, vol. 20, no. 04, pp. 293–318, 2012.
KAISER, ALEXANDER, et al. “COUPLED SINGULAR VALUE DECOMPOSITION OF A CROSS-COVARIANCE MATRIX.” International Journal of Neural Systems, vol. 20, no. 04, World Scientific Pub Co Pte Lt, 2012, pp. 293–318, doi:10.1142/S0129065710002437.

Export

Markierte Publikationen

Open Data LibreCat

Suchen in

Google Scholar