Nikita Kalinin

Institute of Science and Technology Austria.

I am a PhD student at the Institute of Science and Technology Austria (ISTA), advised by Christoph Lampert.

My research focuses on the theoretical foundations of differential privacy, particularly matrix factorization mechanisms for private machine learning and continual observation. I enjoy designing algorithms and studying their theoretical properties. More broadly, I am interested in connections between privacy, algorithms, statistics, learning theory, and optimization.

News

Nov 19, 2026 Received the Google PhD Fellowship in Privacy, Safety, and Security!
Nov 06, 2026 Gave a talk on sampling-free privacy accounting at the University of Copenhagen. Thank you, Amartya, for hosting me!
Oct 30, 2026 Gave a talk on pure-DP continual counting at Aarhus University. Thank you, Kasper, for hosting me!
Sep 29, 2026 Our paper on pure-DP continual counting, joint work with Pavel Arkhipov, was accepted to SODA 2027—my first TCS conference paper!
Sep 24, 2026 Our paper on sampling-free privacy accounting, with Jan Schuchardt, was accepted to NeurIPS 2026. Looking forward to the trip to Sydney!

Selected Publications

For the full list of publications, see my Google Scholar profile.

  1. NeurIPS
  2. FORC
    Nikita P. Kalinin, Joel Daniel Andersson
    FORC 2026
  3. ICLR
    Nikita P. Kalinin, Ryan McKenna, Jalaj Upadhyay, Christoph H. Lampert
    ICLR 2026
  4. NeurIPS
    Nikita P. Kalinin, Jalaj Upadhyay, Christoph H. Lampert
    NeurIPS 2025
  5. AISTATS
    Nikita P. Kalinin, Lukas Steinberger
    AISTATS 2025
  6. NeurIPS
    Nikita P. Kalinin, Christoph H. Lampert
    NeurIPS 2024