Online Program Home
My Program

Abstract Details

Activity Number: 636 - Graphical Models: From Foundations to Applications
Type: Invited
Date/Time: Thursday, August 1, 2019 : 10:30 AM to 12:20 PM
Sponsor: IMS
Abstract #308023
Title: Minimax Prediction in Tree Ising Models
Author(s): Guy Bresler*
Companies: Massachusetts Institute of Technology (MIT)

Graphical models are often used to facilitate efficient computation of posteriors in order to make predictions. With this objective in mind, we consider the problem of estimating the parameters of a graphical model with known structure from samples such that posteriors computed using the model are accurate. Focusing on tree-structured binary Markov random fields, we prove a sharp characterization of accuracy of size k marginals (and hence posteriors conditioned on k-1 variables) in terms of number of samples. Joint work with Mina Karzand.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program