GSO Spring Speaker 2016

Probabilistic Modeling of Big Table and Networks

David B. Dunson
Arts and Sciences Distinguished Professor
Departments of Statistical Science, Mathematics, and Electrical & Computer Engineering, Duke University

Start Date and Time: Fri, 22 Apr 2016, 10:30 AM

End Date and Time: Fri, 22 Apr 2016, 11:20 AM

Venue: BRNG 2290


In applications, data consist of high-dimensional complex and highly-structured discrete data. Our focus here is on high-dimensional unordered categorical data, which arise in epidemiology, social surveys and brain connectomics. In the first part of the talk, I will focus on data that can be structured as a multiway contingency table but otherwise have no obvious structure a priori. For such problems, we rely on probabilistic tensor factorizations, introducing new classes of factorizations, discussing relationships with sparse log-linear models, sketching theory on rates of convergence, and considering applications in social science surveys and genomics. In the second part of the talk, I focus on the case in which the categorical data consist of indicators of connections between pairs of nodes in a network, motivated in particular by brain connectomic studies. The probability distribution for such network-valued random variables can be conveniently represented via a hierarchical latent space representation. We propose a Bayesian approach to inference and show exciting results in performing inferences on differences in brain structure with phenotypes.

Last Updated: Sep 19, 2017 8:29 AM

Purdue Department of Statistics, 250 N. University St, West Lafayette, IN 47907

Phone: (765) 494-6030, Fax: (765) 494-0558

© 2018 Purdue University | An equal access/equal opportunity university | Copyright Complaints

Trouble with this page? Disability-related accessibility issue? Please contact the College of Science Webmaster.