David B. Dunson 
Department of Statistical Science
Duke University
Click to view the Hartley Program.
“Bayes for Big Tables and Networks”
This talk focuses on the challenging problems of analyzing high-dimensional categorical or network-structure data without restrictive parametric assumptions. There is abundant applied motivation drawn from genomics and neurosciences. A general approach is proposed based on probabilistic low rank factorizations, some theory support is provided (without details) and we illustrate the methods.
