AIR Distinguished Speaker Series: Andrew Gordon Wilson, Associate Professor, New York University

Date: March 6, 2024
Time: 11:00am – 12:00pm ET
Location: Center for Computing & Data Sciences, 665 Commonwealth Ave, Room 1101 (11th floor), Boston, MA 02215
***To access the higher levels, please use the elevators by Saxby’s.
Register

Speaker: Andrew Gordon Wilson, Associate Professor at the Courant Institute of Mathematical Sciences and Center for Data Science, New York University 

Talk Title: How do we build models that learn and generalize?

Abstract: In this talk, Prof. Wilson will present a philosophy for model construction, grounded in probability theory. He will exemplify this approach with methods that exploit loss surface geometry for scalable and practical Bayesian deep learning, and resolutions to seemingly mysterious generalization behavior such as overparametrization, double descent, and benign overfitting.

Biography: Andrew Gordon Wilson is an Associate Professor at the Courant Institute of Mathematical Sciences and Center for Data Science at NYU. Prof. Wilson wishes to develop a prescriptive foundation for building intelligent autonomous systems, with work involving Bayesian inference, distribution shifts, scientific discovery, and generalization in deep learning. He has been Workshop Chair, Tutorial Chair, EXPO Chair, and Senior Area Chair for major machine learning conferences, and has received numerous awards, including the NSF CAREER Award, the Amazon Research Award, and best paper, reviewer, area chair, and dissertation awards.

Faculty Host: Brian Kulis, Associate Professor, Engineering (ECE, SE)