Dear All,
We are pleased to invite you to the IEOR Seminar by Parthe Pandit, C-MInDS, IIT Bombay. The details are given below.
Title: Precise asymptotic analysis of some high dimensional statistical inverse problems
Speaker: Parthe Pandit, C-MInDS, IIT Bombay.
Date & time: 26th March.2026, 2:30 pm
Venue: IEOR 211 (Seminar Room), 2nd floor, IEOR building
Abstract: Estimation via regularized empirical risk minimization is central in modern data science. In high-dimensional random linear models, Approximate Message Passing (AMP) provides an algorithmic framework that admits a precise asymptotic characterization of estimator performance, without unknown constants. We introduce AMP, relate it to proximal and operator-splitting based methods (e.g., ADMM), and sketch the main ideas behind the asymptotic convergence result.
Speaker Bio: Parthe Pandit is a core faculty member in C-MInDS at IIT Bombay.
Host: N. Hemachandra