Prateeti Mukherjee

EnCORE 4501, Atkinson Hall

Hi! I’m a third-year Ph.D. student at UC San Diego, where I’m extremely fortunate to be advised by Arya Mazumdar.

I work on statistical learning theory, mostly on high-dimensional estimation from coarse or unreliable observations. I’m drawn to fundamental questions that are grounded in practical phenomena, like when simple algorithms work for reasons we cannot yet explain, or when a problem seems hard in general but is tractable on the instances we care about.

Before grad school, I had the privilege of spending three years at Microsoft Research India, working with Satya Lokam on privacy-preserving machine learning.

Please feel free to reach out at pmukherjee[at]ucsd[dot]edu

Prateeti Mukherjee

Selected papers

Following the convention in theoretical computer science, authors are listed alphabetically by last name.

  • On the Role of Normalization in Binary Iterative Hard Thresholding for 1-bit Compressed Sensing
    Arya Mazumdar and Prateeti Mukherjee
    ACM–SIAM Symposium on Discrete Algorithms (SODA), 2027
    Paper
  • Rare-Tail Statistics For Learning Biased Gaussian Halfspaces with Label Noise
    Arya Mazumdar, Prateeti Mukherjee, and Harsh Vardhan
    Advances in Neural Information Processing Systems (NeurIPS), 2026
    Paper
  • On the Query Complexity of Training Data Reconstruction in Private Learning
    Satya Lokam and Prateeti Mukherjee
    Preprint, 2023
    Paper

For a complete list, see my Google Scholar profile.