Research

My research interests lie broadly at the intersection of time series and learning theory, with a central focus on developing valid statistical inference methods for data exhibiting complex dependency structures. Most recently, I've been exploring the emerging statistical foundations of large language models.

Preprints

  • Stability beyond bounded differences: sharp generalization bounds under finite Lp moments.
    Q. Lei, S. Bonnerjee, Y. Han, W. B. Wu
    Under review.
  • Identifying stability regions of SGD with constant learning rates.
    O. Goldreich, S. Bonnerjee, Q. Lei, J. Li, W. B. Wu
    Under review.
  • Fast segmentation of watermarked texts from large language models through epidemic change-points framework.
    S. Bonnerjee, S. Karmakar, S. Roy
    Under Review at JASA, 2025
  • How Private is Your Attention? Bridging Privacy with In-Context Learning.
    arXiv
    S. Bonnerjee, Z. W. Yeon, A. Asch, S. Nandy, P. Ghosal
    Under Review at TMLR, 2025

Journal Publications

  • Testing synchronization of change-points for multiple time series.
    S. Bonnerjee, S. Karmakar, M. Cheng, W. B. Wu
    Major Revision at Biometrika, 2025
    🏆 IMS Hannan Graduate Student Travel Award, 2025
  • Gaussian Approximation For Non-stationary Time Series with Optimal Rate and Explicit Construction.
    arXiv
    S. Bonnerjee, S. Karmakar, W. B. Wu
    Annals of Statistics, 52(5): 2293–2317, 2024
  • A Generalized Epidemiological Model with Dynamic and Asymptomatic Population.
    arXiv
    A. Ghatak, S. S. Patel, S. Bonnerjee, S. Roy
    Statistical Methods in Medical Research, 31(11): 2137–2163, 2022

Conference Publications

  • Sharp asymptotic theory for Q-learning with LD2Z learning rate and its generalization.
    S. Bonnerjee, Z. Lou, W. B. Wu
    ICLR 2026, Poster
  • Sharp Gaussian approximations for Decentralized Federated Learning.
    NeurIPS arXiv
    S. Bonnerjee, S. Karmakar, W. B. Wu
    NeurIPS 2025, Spotlight (Top 3%)
  • Asymptotic theory of SGD with a general learning-rate.
    NeurIPS
    O. Goldreich, Z. Wei, S. Bonnerjee, J. Li, W. B. Wu
    NeurIPS 2025, Poster

In Preparation

  • Inference for spatial random effects model under dependence.
    S. Bonnerjee, S. Deb, W. B. Wu
    To be submitted at Biometrika, 2025
  • Stable convergence of Stochastic Gradient Descent for non-convex objectives.
    S. Bonnerjee, Y. Han, W. B. Wu
    2026