Research Interests

Preprints

(2025) Testing synchronization of change-points for multiple time series. [Preprint]
Major Revision at Biometrika
S. Bonnerjee, S. Karmakar, M. Cheng, and W. B. Wu.
IMS Hannan Graduate Student Travel Award, 2025.
(2025) Identifying stability regions of SGD with constant learning rates. [Preprint]
To be submitted at Journal of American Statistical Association (JASA)
O. Goldreich, S. Bonnerjee, Q. Lei, J. Li, and W. B. Wu.

(2025) WISER: Segmenting watermarked region - an epidemic change-point perspective. [ArXiv]
S. Bonnerjee, S. Karmakar, and S. Roy.

(2025) How Private is Your Attention? Bridging Privacy with In-Context Learning. [ArXiv]
S. Bonnerjee, Z. W. Yeon, A. Asch, S. Nandy, and P. Ghosal.
(2025) Sharp asymptotic theory for Q-learning with LD2Z learning rate and its generalization. [Preprint]
S. Bonnerjee, Z. Lou, and W. B. Wu.
(2025) Stable convergence of Stochastic Gradient Descent for non-convex objectives. [Preprint]
Under Review at PNAS
S. Bonnerjee, Y. Han, and W. B. Wu.

(2024) Gaussian Approximation For Non-stationary Time Series with Optimal Rate and Explicit Construction.[Journal] [ArXiv]
Annals of Statistics, 52(5): 2293–2317.
S. Bonnerjee, S. Karmakar, and W. B. Wu.
(2022) A Generalized Epidemiological Model with Dynamic and Asymptomatic Population.[Journal] [ArXiv]
Statistical Methods in Medical Research, 31(11): 2137–2163.
A. Ghatak, S. S. Patel, S. Bonnerjee, and S. Roy.

(2025) Sharp Gaussian approximations for Decentralized Federated Learning. [ArXiv]
Accepted at NeurIPS 2025, Main Conference Track, Spotlight (Top 3%)
S. Bonnerjee, S. Karmakar, and W. B. Wu.
(2025) Asymptotic theory of SGD with a general learning-rate. [Preprint]
Accepted at NeurIPS 2025, Main Conference Track, Poster
O. Goldreich, Z. Wei, S. Bonnerjee, J. Li, and W. B. Wu.