Abstract
Joint SVD is a problem of finding a pair of unitary matrices which simultaneously diagonalizes several (possibly non-square) matrices. This paper compares two main approaches to joint SVD problem, “approach via joint diagonalization” and “direct approach”. The former is relatively easy to implement because we can make use of literature of joint diagonalization algorithms while the latter has advantages in numerical accuracy and flexibility to fit on-line applications. Numerical simulation for comparison using gradient-based algorithms verifies that the latter has advantage in numerical accuracy.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Cardoso, J.-F., Souloumiac, A.: Blind beamforming for non Gaussian signals. IEEE Proc. -F 140, 362–370 (1993)
Cardoso, J.-F., Souloumiac, A.: Jacobi angles for simultaneous diagonalization. SIAM J. Mat. Anal. Appl. 17, 161–164 (1996)
Molgedey, L., Schuster, H.G.: Separation of a mixture of independent signals using time delayed correlations. Phys. Rev. Lett. 72, 3634–3637 (1994)
Pesquet-Popescu, B., Pesquet, J.-C., Petropulu, A.P.: Joint singular value decomposition - a new tool for separable representation of images. In: Proc. 2001 International Conference on Image Processing, vol. 2, pp. 569–572 (2001)
Hori, G.: A new approach to joint diagonalization. In: Proc. 2nd Int. Workshop on Independent Component Analysis and Blind Signal Separation (ICA 2000), Helsinki, Finland, pp. 151–155 (2000)
Hori, G.: A general framework for SVD flows and joint SVD flows. In: Proc. 2003 IEEE International Conference on Acoustics, Speech and Signal Processing, vol. 2, pp. 693–696 (2003)
Tanaka, T., Fiori, S.: Least squares approximate joint diagonalization on the orthogonal group. In: Proc. 2007 IEEE International Conference on Acoustics, Speech and Signal Processing, vol. 2, pp. 649–652 (2007)
Fiori, S.: Singular value decomposition learning on double Stiefel manifold. Intl. J. Neural Systems 13(2), 155–170 (2003)
Helmke, U., Moore, J.B.: Optimization and dynamical systems. Addison-Wesley, Reading (1994)
Helmke, U., Moore, J.B.: Singular value decomposition via gradient and self equivalent flows. Linear Algebra Appl. 69, 223–248 (1992)
Chu, M.T., Driessel, K.R.: The projected gradient method for least squares matrix approximations with spectral constraints. SIAM J. Numer. Anal. 27, 1050–1060 (1990)
Smith, S.T.: Dynamical systems that perform the singular value decomposition. System and Control Letters 16, 319–328 (1991)
Manton, J.H.: Optimisation algorithms exploiting unitary constraints. IEEE Trans. Signal Processing 50, 635–650 (2002)
Edelman, A., Arias, T.A., Smith, S.T.: The geometry of algorithms with orthogonality constraints. SIAM J. Matrix Anal. Appl. 20, 303–353 (1998)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Hori, G. (2009). Comparison of Two Main Approaches to Joint SVD. In: Adali, T., Jutten, C., Romano, J.M.T., Barros, A.K. (eds) Independent Component Analysis and Signal Separation. ICA 2009. Lecture Notes in Computer Science, vol 5441. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-00599-2_6
Download citation
DOI: https://doi.org/10.1007/978-3-642-00599-2_6
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-00598-5
Online ISBN: 978-3-642-00599-2
eBook Packages: Computer ScienceComputer Science (R0)