PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

High Dimensional Discriminant Analysis
Charles Bouveyron, Stephane Girard and Cordelia Schmid
Communications in Statistics : Theory and Methods Volume 36, Number 14, pp. 2607-2623, 2007.

Abstract

We propose a new discriminant analysis method for high-dimensional data, called High-Dimensional Discriminant Analysis (HDDA). Our approach is based on the assumption that high-dimensional data live in different subspaces with low dimensionality. We therefore propose a new parameterization of the Gaussian model which combines the ideas of dimension reduction and constraints on the model. This parameterization takes into account the specific subspace and the intrinsic dimension of each class to limit the number of parameters to estimate. In addition, it is possible to make additional assumptions on the model to further limit the number of parameters. Our experiments on artificial and real datasets highlight that HDDA is more efficient than classical methods in high-dimensional spaces and with small learning datasets.

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
ID Code:3885
Deposited By:Charles Bouveyron
Deposited On:25 February 2008