No CrossRef data available.
Article contents
Gradient algorithms for principal component analysis
Published online by Cambridge University Press: 17 February 2009
Abstract
Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the 'Save PDF' action button.
The problem of principal component analysis of a symmetric matrix (finding a p-dimensional eigenspace associated with the largest p eigenvalues) can be viewed as a smooth optimization problem on a homogeneous space. A solution in terms of the limiting value of a continuous-time dynamical system is presented. A discretization of the dynamical system is proposed that exploits the geometry of the homogeneous space. The relationship between the proposed algorithm and classical methods are investigated.
Information
- Type
- Research Article
- Information
- Copyright
- Copyright © Australian Mathematical Society 1996
References
[1]Ammar, G. and Martin, C., “The geometry of matrix eigenvalue methods”, Acta Applicandae Mathematicae 5 (1986) 239–278.CrossRefGoogle Scholar
[2]Bloch, A. M., “A completely integrable Hamiltonian system associated with line fitting in complex vector spaces”, Bull. Amer. Math. Soc. 12 (1985) 250–254.CrossRefGoogle Scholar
[3]Brockett, R. W., Differential geometry and the design of gradient algorithms, Vol. 54 Proc. Symp. Pure Mathematics (1993) 69–92.CrossRefGoogle Scholar
[4]Byrnes, C. I. and Willems, J. C., “Least-squares estimation, linear programming and momentum: A geometric parametrization of local minima”, IMA J. Math. Control and Information 3 (1986) 103–118.CrossRefGoogle Scholar
[5]Chu, M. T., “On the continuous realization of iterative processes”, SIAM Review 30 (1988) 375–387.CrossRefGoogle Scholar
[6]Deift, P., Nanda, T. and Tomei, C., “Ordinary differential equations for the symmetric eigenvalue problem”, SIAM J. ofNumer. Anal. 20 (1983) 1–22.CrossRefGoogle Scholar
[7]Duistermaat, J. J., Kolk, J. A. C. and Varadarajan, V. S., “Functions, flows and oscillatory integrals on flag manifolds and conjugacy classes in real semi-simple Lie groups”, Composito Mathematica 49 (1983) 309–398.Google Scholar
[8]Faddeev, D.K. and Faddeeva, V. N., Computational methods of linear algebra (W. H. Freeman and Co., San Francisco, 1963).Google Scholar
[9]Gibson, C. G., Singular points of smooth mappings, Vol. 25 Res. Notes in Math. (Pitman, London, United Kingdom, 1979).Google Scholar
[10]Golub, G. H. and Loan, C. F. Van, Matrix computations (The Johns Hopkins University Press, Baltimore, Maryland U.S.A., 1989).Google Scholar
[11]Helmke, U. and Moore, J. B., Optimization and dynamical systems, Communications and Control Engineering (Springer-Verlag, London, 1994).CrossRefGoogle Scholar
[12]Horn, R. A. and Johnson, C. R., Matrix analysis (Cambridge University Press, Cambridge, U.K., 1985).CrossRefGoogle Scholar
[13]Mahony, R. E., “Optimization algorithms on homogeneous spaces: with aplications in linear systems theory”, Ph. D. Thesis, Department of Systems Engineering, Canberra, Australia, 1994.Google Scholar
[14]Moore, J. B., Mahony, R. E. and Helmke, U., “Numerical gradient algorithms for eigenvalue and singular value calculations”, SIAM J. Matrix Analysis (1994).CrossRefGoogle Scholar
[15]Oja, E., “A simplified neuron model as a principal component analyzer”, J. Math. Biology 15 (1982) 267–273.CrossRefGoogle ScholarPubMed
[16]Oja, E., “Neural networks, principal components, and subspaces”, Intern. J. Neural Systems 1. (1989) 61–68.CrossRefGoogle Scholar
[17]Warner, F. W., Foundations of differentiable manifolds and Lie groups, Graduate texts in Mathem- atics (Springer-Verlag, New York, U.S.A., 1983).CrossRefGoogle Scholar
[18]Yan, W.-Y., Helmke, U. and Moore, J. B., “Global analysis of Oja's flow for neural networks”, IEEE Tran. Neural Networks 5 (1994), 674–683.Google ScholarPubMed
You have
Access
- 12
- Cited by