Expert in machine learning theory, Jason Lee, wins funding from the Navy's Young Investigator Program

Written by
Scott Lyon
May 13, 2021

The Office of Naval Research has awarded Assistant Professor of Electrical and Computer Engineering Jason Lee three years of funding through its Young Investigator Program. The award covers a range of critical expenses, including the salaries and stipends of researchers in his lab.

Lee studies the theoretical foundations of machine learning and artificial intelligence, especially the architectures of deep learning models. Drawing from a variety of quantitative fields, including statistics and optimization, Lee has established himself as an authority on the mathematics of deep learning. His work seeks methods to improve the speed and guarantee the efficacy of such models, even when those models are trained on low-quality data, as is most often the case in real world applications. With this award he will contribute to the Navy's mathematical data science project through the development of theory guiding stochastic gradient descent, a method for finding high performance algorithms.

This year's 38 Young Investigator Program awardees, selected from 260 applicants, will share a total $20 million in funding, according to an official statement. Three of those awardees are Princeton University faculty members, including Lee; Leslie Schoop, assistant professor of chemistry; and Sanfeng Wu, assistant professor of physics. All of the candidates were tenure-track researchers who received a Ph.D. after 2013.

Lee received his Ph.D. from Stanford University in 2015. He was an assistant professor for three years at the University of Southern California, then spent one year at the Institute for Advanced Study as part of the theoretical machine learning group. In 2019 he was awarded a two-year Sloan Research Fellowship in computer science. He joined Princeton that same year, and was awarded a teaching commendation last year for his graduate-level course "Foundations of Deep Learning."