Session 4: Conditional Mixture Modeling and Model-based Clustering

Presenter Information/ Coauthors Information

Yang Wang, College of CharlestonFollow

Presentation Type

Invited

Student

No

Track

Tools

Abstract

Due to a potentially high number of parameters, finite mixture models are often at the risk of overparameterization even for a moderate number of components. This can lead to overfitting individual components and result in mixture order underestimation. One of the most popular approaches to address this issue is to reduce the number of parameters by considering parsimonious models. The vast majority of techniques in this direction focus on the reparameterization of covariance matrices associated with mixture components. We propose an alternative approach that emphasizes modeling cluster locations. The developed procedure enjoys remarkable modeling flexibility, especially noticeable in the presence of non-compact clusters. Due to an attractive closed form formulation, speedy parameter estimation is available by means of the EM algorithm. The utility of the proposed method is illustrated on synthetic and well-known classification data sets.

Start Date

2-8-2022 9:50 AM

End Date

2-8-2022 10:50 AM

This document is currently not available here.

Share

COinS
 
Feb 8th, 9:50 AM Feb 8th, 10:50 AM

Session 4: Conditional Mixture Modeling and Model-based Clustering

Herold Crest 253 C

Due to a potentially high number of parameters, finite mixture models are often at the risk of overparameterization even for a moderate number of components. This can lead to overfitting individual components and result in mixture order underestimation. One of the most popular approaches to address this issue is to reduce the number of parameters by considering parsimonious models. The vast majority of techniques in this direction focus on the reparameterization of covariance matrices associated with mixture components. We propose an alternative approach that emphasizes modeling cluster locations. The developed procedure enjoys remarkable modeling flexibility, especially noticeable in the presence of non-compact clusters. Due to an attractive closed form formulation, speedy parameter estimation is available by means of the EM algorithm. The utility of the proposed method is illustrated on synthetic and well-known classification data sets.