Texas A&M - Labors of Lab
This installment of "Labors of Lab" features statistics Ph.D candidate Richard Payne '17, who is using Bayesian statistics -- specifically big-data classification methods -- in collaboration with Texas A&M distinguished professor of statistics Bani Mallick to make algorithms more efficient.
Captivated by the power of data, Texas A&M graduate student Minsuk Shin '16 is pairing his innate skill with numbers and probability with Bayesian statistics to unlock countless secrets about the world and potentially impact industries from business and government to technology and medicine.Read More →
Texas A&M Statistics graduate Ersen Arseven '74 recently commemorated the 15-year anniversary of his wife Susan M. Arseven's untimely death by establishing his most recent memorial endowment at Texas A&M in her honor, the Susan M. Arseven '75 Chair in Data Science and Computational Statistics.Read More →
05:00 PM / 07:30 PM StataCorp Park Pavilion 979-845-3141
11:30 AM / 12:30 PM Blocker Building (BLOC), Room 113 979-845-3141
Department of Statistics
University of Michigan
"An Optimal Transport Based Theory of Inference with Hierarchical Models"
Hierarchical models present a powerful tool in statistics and numerous applied fields. They allow for statistical dependence to be conveniently expressed via latent variables. They also enable the "borrowing of strength" --- a heuristic argument hinting at the benefit of combining multiple data sets by making them share some latent variables in common. To understand this phenomenon, one needs to understand the convergence behavior of latent variables that arise in hierarchical models. In this talk I will describe some progress in our study of such questions. We will start with classic models such as location-scale Gaussian mixtures, shape-scale Gamma mixtures, and establish new results on rates of convergence of mixing distributions arising from such models. We will then present results for several hierarchical models widely used in Bayesian nonparametrics, including the Dirichlet process mixtures, and the hierarchical Dirichlet processes. The main tool in our theory is the use of a class of optimal transport based distances (i.e., Wasserstein distances) of probability measures. They turn out to be particularly natural for analyzing the geometry of mixture and hierarchical models.
11:00 AM / 01:00 PM Rudder Tower, Room 301 979-845-3141
SETCASA is the local chapter of the American Statistical Association. A Poster Session is scheduled for Friday, October 16, 2015 from 11:00 am to 1:00 pm in Rudder Tower, Room 301. Statistics Faculty and Students are welcome and encouraged to attend.
To present a poster, please contact Dr. Matthias Katzfuss at firstname.lastname@example.org.