publications

publications by categories in reversed chronological order.

2026

  1. ICLR
    Back to square roots: An optimal bound on the matrix factorization error for multi-epoch differentially private SGD
    Nikita P Kalinin ,  Ryan McKenna ,  Jalaj Upadhyay , and 1 more author
    International Conference on Learning Representations (ICLR 2026), Apr 2026
  2. arXiv
    DP-lambdaCGD: Efficient Noise Correlation for Differentially Private Model Training
    Nikita P Kalinin ,  Ryan McKenna ,  Rasmus Pagh , and 1 more author
    arXiv, Jan 2026
  3. arXiv
    Sampling-Free Privacy Accounting for Matrix Mechanisms under Random Allocation
    Jan Schuchardt ,  and  Nikita P Kalinin
    arXiv, Jan 2026
  4. arXiv
    Matrix Factorization for Practical Continual Mean Estimation Under User-Level Differential Privacy
    Nikita P Kalinin ,  Ali Najar ,  Valentin Roth , and 1 more author
    arXiv, Jan 2026

2025

  1. AIRoV
    DP-KAN: Differentially Private Kolmogorov-Arnold Networks
    Nikita P Kalinin ,  Simone Bombari ,  Hossein Zakerinia , and 1 more author
    The Austrian Symposium on AI and Vision (AIRoV 2025), Jul 2025
  2. arXiv
    Binned group algebra factorization for differentially private continual counting
    Monika Henzinger ,  Nikita P Kalinin ,  and  Jalaj Upadhyay
    arXiv, Apr 2025
  3. FORC
    Normalized square root: Sharper matrix factorization bounds for differentially private continual counting
    Monika Henzinger ,  Nikita P Kalinin ,  and  Jalaj Upadhyay
    Symposium on the Foundations of Responsible Computing (FORC 2026), Sep 2025
  4. FORC
    Learning Rate Scheduling with Matrix Factorization for Private Training
    Nikita P Kalinin ,  and  Joel Daniel Andersson
    Symposium on the Foundations of Responsible Computing (FORC 2026), Nov 2025
  5. NeurIPS
    Continual Release Moment Estimation with Differential Privacy
    Nikita P Kalinin ,  Jalaj Upadhyay ,  and  Christoph H Lampert
    Conference on Neural Information Processing Systems, Jun 2025

2024

  1. AISTATS
    Efficient Estimation of a Gaussian Mean with Local Differential Privacy
    Nikita Kalinin ,  and  Lukas Steinberger
    Artificial Intelligence and Statistics (AISTATS) 2025, Feb 2024
  2. NeurIPS
    Banded square root matrix factorization for differentially private model training
    Kalinin Nikita ,  and  Lampert Christoph
    Advances in Neural Information Processing Systems (NeurIPS) 2024, Dec 2024