Bayesian mixtures of multiple scale distributions

Florence Forbes (INRIA Grenoble Rhone-Alpes)

Friday 15th March 15:00-16:00 Maths 311B

Abstract

Multiple scale distributions are multivariate distributions that exhibit a variety of shapes not necessarily elliptical while remaining analytical and tractable. In this work we consider mixtures of such distributions for their ability to handle non standard typically non-gaussian clustering tasks. We propose a Bayesian formulation of the mixtures and a tractable inference procedure based on  variational approximation. The interest of such a Bayesian formulation is illustrated on an important mixture model selection task, which is the issue of selecting automatically the number of components. We derive promising procedures that can  be carried out in a single-run, in contrast to the more costly comparison of information criteria.

Add to your calendar

Download event information as iCalendar file (only this event)