Algorithms for Regularized Linear Discriminant Analysis

Authors: Jan Kalina, Jurjen Duintjer Tebbens
Publication year: 2015
Publication type: Konferenční příspěvek (mezinárodní konf.)
Periodical: BIOINFORMATICS 2015. Proceedings of the International Conference on Bioinformatics Models, Methods and Algorithms
Projects: Constructing Advanced Comprehensible Classifiers
Iterative Methods in Computational Mathematics: Analysis, Preconditioning, and Applications
Bibliographic citation: KALINA, Jan, DUINTJER TEBBENS, Jurjen. Algorithms for Regularized Linear Discriminant Analysis. In: BIOINFORMATICS 2015. Proceedings of the International Conference on Bioinformatics Models, Methods and Algorithms. Lisbon: Scitepress, 2015, S. 128-133. ISBN 978-989-758-070-3
Dostupné z: doi:10.5220/0005234901280133
DOI: 10.5220/0005234901280133
ID: 439199

Abstract

This paper is focused on regularized versions of classification analysis and their computation for highdimensional data. A variety of regularized classification methods has been proposed and we critically discuss their computational aspects. We formulate several new algorithms for regularized linear discriminant analysis, which exploits a regularized covariance matrix estimator towards a regular target matrix. Numerical linear algebra considerations are used to propose tailor-made algorithms for specific choices of the target matrix. Further, we arrive at proposing a new classification method based on L2-regularization of group means and the pooled covariance matrix and accompany it by an efficient algorithm for its computation.