Towards Building the Mathematical Foundations of Data Science

Abstract

In this talk, I will discuss my view of what the mathematical foundations of data science should entail by giving some examples from my work of theories that put real-world considerations/data at the forefront. In particular, I will present 1) the statistical algorithms framework, which helps to explain the difficulty of many problems, including community-detection; 2) new results in adaptive data analysis, which give methods for reusing data without overfitting; and 3) some interesting machine learning reductions, which can provide relatively simple methods to get principled algorithms, including for bandit/reinforcement learning.

Bio:

Lev Reyzin is a Professor of Mathematics, Statistics, and Computer Science and the Site Director of the Institute for Data, Econometrics, Algorithms, and Learning at the University of Illinois at Chicago. Prior to UIC, Lev was a Simons Postdoctoral Fellow at Georgia Tech, and before that, an NSF CI-Fellow at Yahoo! Research. Lev received his Ph.D. on an NSF graduate research fellowship from Yale under Dana Angluin and his undergraduate degree from Princeton. He serves as the Treasurer of the Association for Algorithmic Learning Theory, as an Associate Editor of Ann. Math. Artif. Intell., and as an Editorial Board Reviewer for J. Mach. Learn. Res. Recently, he was program chair for ALT 2017 and ISAIM 2020. His work has earned awards at ICML, COLT, and AISTATS and has been funded by individual grants from the NSF and DOD.