Computational & Applied Mathematics Seminar

August 27, 2021 - 1:00 PM to 2:00 PM

Speaker: Pedram Rooshenas
Date and Time: Aug 27, 2021, 1:00 p.m. -2:00 p.m (Fretwell 315)
Title: Energy-Based Models: Promises and Challenges
Abstract: Recent advances in deep neural networks (DNNs) have revolutionized fields such as natural language processing, computer vision, and robotics, while also recently impacting other fields such as biology, computational mechanics, and health care. Many problems in these domains require a joint model to describe the correlation among a set of variables, such as the correlation of words in a translated sentence. Classically, these correlations were modeled using graphical models (such as Markov random fields) defined over the variables. This talk will show how we can leverage DNNs to describe these correlations using energy-based models (EBMs). I will also show how we can use EBMs for different problem settings and discuss their promises and challenges.