Niladri S. Chatterji

PhD Student

University of California Berkeley

I am a fifth year PhD student at UC Berkeley. I am associated with the Berkeley Artificial Intelligence Research group (BAIR). My research interests lie at the intersection of Theoretical Statistics and Machine Learning. Some areas I am interested in and have worked on in the past include, understanding the generalization ability of overparameterized classifiers, Monte Carlo methods, optimization guarantees for structured non-convex problems and online learning with partial information (bandit feedback). I am fortunate to be advised by Peter Bartlett and Michael DeWeese.

I graduated from Indian Institute of Technology Bombay in 2015 with a dual degree B.Tech+M.Tech in Engineering Physics.


  • Statistical learning theory
  • Understanding deep neural networks
  • Online learning
  • Markov chain Monte Carlo methods


  • PhD in Physics, 2021 (expected)

    University of California Berkeley

  • BTech and MTech in Engineering Physics, 2015

    Indian Institute of Technology Bombay

Selected Publications

The intriguing role of module criticality in the generalization of deep networks

ICLR 2020; also appeared at Workshops on ML with Guarantees & on Science of Deep Learning, NeurIPS 2019.