Online Program Home
My Program

Abstract Details

Activity Number: 385 - Leo Breiman Award
Type: Invited
Date/Time: Tuesday, July 30, 2019 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistical Learning and Data Science
Abstract #300473
Title: Restricted Boltzmann Machines and Truncated Gaussian Distributions
Author(s): Yichao Wu*
Companies: The University of Illinois at Chicago
Keywords: Boltzmann machines; truncated Gaussian; high dimension

A latent Gaussian model is commonly used for restricted Boltzmann machines. The corresponding estimation scheme typically requires calculating the mean and covariance matrix of truncated Gaussian distributions. In this talk, we will present a new efficient algorithm to evaluate probabilities and expectations with respected to a truncated multivariate Gaussian distributions and illustrate its use in the estimation of restricted Boltzmann machines.

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

Back to the full JSM 2019 program