Harvard University homepage
HARVARD.EDU

Research

Here is a link to my CV. See also Google Scholar. I gratefully thank NSF (DMS CAREER 2239234), ONR (N00014-23-1-2489) and AFOSR (FA9950-23-1-0429) for supporting my research.

Tutorial

  • A Friendly Tutorial on Mean-Field Spin Glass Techniques for Non-Physicists. [ArXiv] [Journal] with Andrea Montanari, 2022.
    This tutorial is based on lecture notes written for a class taught in the Statistics Department at Stanford in the Winter Quarter of 2017. The objective was to provide a working knowledge of some of the techniques developed over the last 40 years by theoretical physicists and mathematicians to study mean field spin glasses and their applications to high-dimenensional statistics and statistical learning.

Publications and Preprints

High-dimensional and Non-parametric Statistics

  • Causal effect estimation under network interference with mean-field methods. [ArXiv] (with Sohom Bhattacharya)- Submitted.
  • On Naive Mean-Field Approximation for high-dimensional canonical GLMs. [ArXiv] (with Sumit Mukherjee and Jiaze Qiu)- Submitted.
  • Understanding Optimal Feature Transfer via a Fine-Grained Bias-Variance Analysis. [ArXiv] (with Yufan Li and Ben Adlam)- Submitted.
  • A Mean Field Approach to Empirical Bayes Estimation in High-dimensional Linear Regression. [ArXiv] (with Sumit Mukherjee and Bodhisattva Sen)- Submitted.
  • Bayes optimal learning in high-dimensional linear regression with network side information. [ArXiv] (with Sagnik Nandy)
    – IEEE Transactions in Information Theory (to appear), 2024+.
  • Random linear estimation with rotationally-invariant designs: Asymptotics at high temperature. [ArXiv] (with Yufan Li, Zhou Fan and Yihong Wu)- IEEE Transactions on Information Theory (to appear), 2023+.
  • Sparse Signal Detection in Heteroscedastic Gaussian Sequence Models: Sharp Minimax Rates. [ArXiv] (with Julien Chhor and Rajarshi Mukherjee)
    Bernoulli (to appear), 2023+.
  • Spectral Universality of Regularized Linear Regression with Nearly Deterministic Sensing Matrices. [ArXiv] (with Rishabh Dudeja and Yue M. Lu)- IEEE Transactions on Information Theory (to appear), 2024+.
  • A New Central Limit Theorem for the Augmented IPW Estimator: Variance Inflation, Cross-Fit Covariance and Beyond. [ArXiv] (with Kuanhao Jiang, Rajarshi Mukherjee and Pragya Sur)- Submitted.
  • High-dimensional Asymptotics of Langevin Dynamics in Spiked Matrix Models. [ArXiv] (with Tengyuan Liang and Pragya Sur)
    Information and Inference (to appear), 2023+.
  • The TAP free energy for high-dimensional linear regression. [ArXiv] (with Jiaze Qiu)
    The Annals of Applied Probability (to appear), 2022+.
  • Regret Minimization in Isotonic, Heavy-Tailed Contextual Bandits via Adaptive Confidence Bands. (with Sabyasachi Chatterjee)– Submitted.
  • Variational Inference in high-dimensional linear regression. [ArXiv] (with Sumit Mukherjee)
    Journal of Machine Learning Research, 2022.
  • On Minimax Exponents of Sparse Testing. [ArXiv] (with Rajarshi Mukherjee)– Submitted.
  • The Overlap Gap Property in Planted Submatrix Recovery. [ArXiv] (with David Gamarnik, Aukosh Jagannath)
    Probability Theory and Related Fields, 181.4(2021):757-814.
  • Optimal Adaptive Inference in Random Design Binary Regression. [ArXiv][Journal] (with Rajarshi Mukherjee)
    Bernoulli, 24.1(2018): 699-739.

Statistical Inference on Networks

Random graphs: typical and atypical properties

Random Combinatorial Optimization, Spin glasses and Universality

Miscellaneous