Systems and Means of Informatics
2021, Volume 31, Issue 3, pp 70-79
COMPUTING BASED ON PROBABILISTIC PRINCIPAL COMPONENT ANALYSIS MODEL
Abstract
An effective solution to problems of multivariate data analysis requires the use of complex probabilistic models, among which the probabilistic model of principal component analysis (PPCA) occupies a worthy place. It allows one to adequately describe real data, formally formulate and solve the problem of choosing its parameters. Practical application of PPCA is associated with a large number of laborious computations. The article discusses techniques for significantly reducing the time spent in calculating the density of the multivariate normal distribution. For this, the expediency of using the Cholesky expansion for the covariance matrix and Woodbury identity for its PPCA factorization is substantiated. A brief description of the experiments is given, which makes it possible to evaluate the real time characteristics of the algorithms and to reveal the conditions for their effective application. Along the way, recommendations are given on the implementation of individual operations.
[+] References (5)
- Krivenko, M.P. 2019. Vybor razmernostey dlya smesi veroyatnostnykh analizatorov glavnykh komponent [Selecting the dimensionality for mixture of probabilistic principal component analyzers]. Sistemy i Sredstva Informatiki - Systems and Means of Informatics 29(3):4-15.
- Tipping, M. E., and C. M. Bishop. 1999. Mixtures of probabilistic principal component analyzers. Neural Comput. 11 (2):443-482.
- Seber, G. A. F., and A. J. Lee. 2003. Linear regression analysis. Hoboken, NJ: Wiley. 557 p.
- Golub, G. H., and C. F. Van Loan. 2013. Matrix computations. 4th ed. Baltimore, MD: The Johns Hopkins University Press. 780 p.
- Yiping, G. Y., and H. D. Bondell. 2010. On robust probabilistic principal component analysis using multivariate t-distributions. Available at: https://arxiv.org/abs/ 2010.10786 (accessed August 26, 2021).
[+] About this article
Title
COMPUTING BASED ON PROBABILISTIC PRINCIPAL COMPONENT ANALYSIS MODEL
Journal
Systems and Means of Informatics
Volume 31, Issue 3, pp 70-79
Cover Date
2021-11-10
DOI
10.14357/08696527210306
Print ISSN
0869-6527
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
PPCA computing; Cholesky factorization; Woodbury matrix identity; computing experience
Authors
M. P. Krivenko 
Author Affiliations
Federal Research Center "Computer Science and Control", Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
|