Learning Hierarchical Models

• Activity Number: 299
• Type: Topic Contributed
• Date/Time: Tuesday, August 6, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Learning and Data Mining
• Author(s): Ruslan Salakhutdinov*+
• Companies: University of Toronto
• Keywords: Graphical models ; Markov Random Fields ; Variational Learning ; MCMC ; Hierarchical Models

Abstract

Extracting meaningful representations from high-dimensional data lies at the core of solving many AI related tasks, including visual object recognition, language and speech perception, anomaly detection, and time series analysis. In this talk I will introduce a broad class of hierarchical graphical models called Deep Boltzmann Machines (DBMs) that contain multiple layers of latent variables. I will describe a new learning algorithm for this class of probabilistic models that uses variational methods and Markov chain Monte Carlo (MCMC). I will then show that DBMs can learn useful hierarchical representations from high-dimensional data, and that they can be successfully applied in many domains, including information retrieval, object recognition, and nonlinear dimensionality reduction.