High-Dimensional Regression with Gaussian Mixtures and Partially-Latent Response Variables

Antoine Deleforge 1 Florence Forbes 2 Radu Horaud 1
1 PERCEPTION - Interpretation and Modelling of Images and Videos
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
2 MISTIS - Modelling and Inference of Complex and Structured Stochastic Systems
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : In this work we address the problem of approximating high-dimensional data with a low-dimensional representation. We make the following contributions. We propose an inverse regression method which exchanges the roles of input and response, such that the low-dimensional variable becomes the regressor, and which is tractable. We introduce a mixture of locally-linear probabilistic mapping model that starts with estimating the parameters of inverse regression, and follows with inferring closed-form solutions for the forward parameters of the high-dimensional regression problem of interest. Moreover, we introduce a partially-latent paradigm, such that the vector-valued response variable is composed of both observed and latent entries, thus being able to deal with data contaminated by experimental artifacts that cannot be explained with noise models. The proposed probabilistic formulation could be viewed as a latent-variable augmentation of regression. We devise expectation-maximization (EM) procedures based on a data augmentation strategy which facilitates the maximum-likelihood search over the model parameters. We propose two augmentation schemes and we describe in detail the associated EM inference procedures that may well be viewed as generalizations of a number of EM regression, dimension reduction, and factor analysis algorithms. The proposed framework is validated with both synthetic and real data. We provide experimental evidence that our method outperforms several existing regression techniques.
Liste complète des métadonnées


https://hal.inria.fr/hal-00863468
Contributeur : Team Perception <>
Soumis le : mardi 18 mars 2014 - 15:16:38
Dernière modification le : mercredi 14 décembre 2016 - 01:07:13
Document(s) archivé(s) le : jeudi 19 juin 2014 - 14:42:10

Fichiers

Identifiants

Citation

Antoine Deleforge, Florence Forbes, Radu Horaud. High-Dimensional Regression with Gaussian Mixtures and Partially-Latent Response Variables. Statistics and Computing, Springer Verlag (Germany), 2015, 25 (5), pp.893-911. <http://link.springer.com/article/10.1007/s11222-014-9461-5>. <10.1007/s11222-014-9461-5>. <hal-00863468v3>

Partager

Métriques

Consultations de
la notice

740

Téléchargements du document

629