Hosted by the College of Engineering and Edwardson School of Industrial Engineering 

Abstract

Statistical prediction is the engine powering dramatic advances in AI: diffusion denoising for generating images, token prediction for large language models, and look-ahead prediction in reinforcement learning. Complex predictive models can behave remarkably well, but can also fail in mysterious and dangerous ways, including biased and mis-calibrated outputs, poor performance under distribution shift, and high uncertainty associated with their predictive performance. This talk discusses these challenges, along with some new approaches for tackling them. We discuss a simple method for combining (potentially misleading) AI-generated data with clean data, and a fast and data-efficient refitting method for assessing predictive performance.

Biography

Martin Wainwright is the Ford Chair Professor in Electrical Engineering and Computer Science and Mathematics at MIT, and affiliated with the Laboratory for Information and Decision Systems and Statistics and the Data Science Center. He joined the MIT faculty in July 2022 after spending 20 years in Statistics and EECS at the University of California, Berkeley.

Martin is broadly interested in machine learning, statistics, information theory and optimization. His work has been recognized by various awards, among them a Guggenheim Foundation Fellowship; the COPSS Presidents’ Award from the Joint Statistical Societies; the David Blackwell Award from the Institute of Mathematical Statistics; selection as a Section Lecturer from the International Congress of Mathematicians; an Alfred P. Sloan Foundation Fellowship; and several best paper awards from the IEEE. He is a Fellow of the IEEE and of the Institute of Mathematical Statistics. He has co-authored several books, including on graphical models, sparse statistical models and high-dimensional statistics.