Annonce

Les commentaires sont clos.

Séminaire du Prof. Geoff McLachlan : "On Some Mixture Distributions for Modelling Complex Datasets"

15 Avril 2015


Catégorie : Autres événements


Nous avons le plaisir de vous inviter au séminaire du Professeur Geoffrey McLachlan, intitulé "On Some Mixture Distributions for Modelling Complex Datasets" et qui aura lieu le jeudi 23 avril de 14h à 16h à l'Université de Toulon, Campus La Garde, Amphi K 018.

Si vous comptez y assister, merci de vous inscrire ici : http://goo.gl/forms/7UBveWvR0T

L'inscription est gratuite mais obligatoire.

Les informations d'accès sont disponibles ici : http://www.univ-tln.fr/Campus-de-La-Garde.html

 

Geoffrey McLachlan est Professeur de Statistique au département de mathématiques à l'Université du Queensland (UQ) en Australie.
Il est actuellement Professeur Visiteur à l'Université de Toulon, laboratoire LSIS.

Plus d'informations sur l'orateur :

Geoff McLachlan, PhD, DSc Professor of Statistics (Personal Chair) in the Department of Mathematics; UQ Vice-Chancellor's Senior Research Fellow Australian Professorial Fellow
Site web : http://www.maths.uq.edu.au/~gjm/
Biographie : https://www.smp.uq.edu.au/node/106/384

Titre du séminaire : On Some Mixture Distributions for Modelling Complex Datasets

Abstract:

Finite mixture distributions are being increasingly used to modelheterogeneous data and to provide a clustering of such data. For multivariate (continuous) data attention has been focussed on normal mixtures due in part to the computational tractability of the normal distribution for multivariate data. In this talk,we consider extensions to the use of the normal distribution forthe components to include the multivariate t-distribution for data with possibly long-tailed clusters.

We also outline recent work on the use of skew versions of the t-distribution to cover situations where the clusters are not elliptically symmetric. Finally, we consider modifications to the standard normal mixture modelfor applications where the number of experimental units n is comparatively small but the underlying dimension p is extremely large, as, for example, in microarray-based genomics and other high-throughput experimental approaches. We consider ways including the use of factor models to reduce the number of parameters in the specification of the component-covariance matrices.

Faicel Chamroukhi, http://chamroukhi.univ-tln.fr
Associate Professor, Computer Science & Statistics
Director of Studies of the BSc degree in Engineering Sciences
University of Toulon - France
Information Sciences and Systems Laboratory (LSIS)- UMR 7296 CNRS
Tel: +33(0)4 94 14 20 06
Fax: +33(0)4 94 14 28 97