Colloquium Series: Rahul Mazumder

Our upcoming event for the Statistics Department Colloquium Series is scheduled for Monday, February 24 from 12:00 – 1:00pm (ET) and will be an in-person presentation Science Center Rm. 316. Lunch will be provided to guests following the talk. This week's speaker will be Rahul Mazumder of the Operations Research and Statistics group at the MIT Sloan School of Management.

Discrete Optimization in Computational Statistics: Linear Models, Foundation Model Compression, and Beyond

Abstract: Many computational tasks frequently encountered in statistics can be formulated as optimization problems involving discrete and continuous variables. Common examples include best subset selection, sparse principal component analysis (PCA), nonparametric additive models, graphical models with sparse precision matrices, and learning permutations, among others. In this talk, I will explore recent algorithmic advancements addressing these problems--with optimality certificates--using tools from discrete optimization (eg, branch-and-bound) and continuous optimization (eg, first order methods in convex optimization) tools rarely used together. I will then discuss how some of these computational statistics tasks arise in the context of compressing large foundation models with billions of parameters (for improved latency) and pruning large black-box tree ensembles (for interpretability purposes). Time permitting, I will also mention applications and deployments of these procedures in collaboration with domain experts.

Rahul Mazumder is the NTU Associate Professor of Operations Research and Statistics at MIT Sloan School of Management. He is affiliated with MIT OR Center, MIT Center for Statistics and Data Science, LIDS and IDSS. His research interests are at the intersection of statistics, machine learning and mathematical programming, and their applications to industry, the government, and the sciences. He is a recipient of the Leo Breiman Junior Award from the American Statistical Association, International Indian Statistical Association Early Career Award in Statistics and Data Science, INFORMS Donald P. Gaver, Jr. Early Career Award for Excellence in Operations Research, INFORMS Optimization Society Young Researchers Prize, Office of Naval Research Young Investigator Award, INFORMS ICS Prize (Honorable Mention). He is/was serving as AE of the Annals of Statistics, Operations Research and Journal of Machine Learning Research, and Bernoulli.