Johanna Düngler

DDSA PhD Fellow
Computer Science, Copenhagen University
GitHub | | LinkedIn

About Me

I am a DDSA PhD Fellow in Machine Learning @ Copenhagen University in the Department of Computer Science, where I am advised by Amartya Sanyal and Rasmus Pagh. My research interest lie in theoretical foundations and guarantees of Machine Learning, particularly in relation to Privacy and Robustness.

Prior to my PhD, I completed my undergraduate education in Mathematics at ETH Zürich. After completing my Master's degree, I worked as a Data Scientist for 2 years at Revolut, developing Machine Learning Models for the Risk Department.

Academic Service & Teaching

Program Membership
P1 Program “Data Privacy in Machine Learning” Pioneer Centre for AI, Denmark.
Local Organizer
IEEE Conference on Secure and Trustworthy Machine Learning (SaTML 2025).
Reviewer
TPDP (2025), AISTATS (2024, 2025).
Teaching Assistant
Privacy in Machine Learning (2025, 2024, University of Copenhagen), Mathematics of Machine Learning (2022, ETH Zürich), Partial Differential Equations (2020, ETH Zürich).

Publications

  • Preprint, 2025

    Beyond Additive Noise: DP for LoRA via Random Projections

    Yaxi Hu, Johanna Düngler, Bernhard Schölkopf, Amartya Sanyal

    PDF

  • NeurIPS, 2025 · TPDP, 2025

    An iterative algorithm for differentially private k-PCA with adaptive noise

    Johanna Düngler, Amartya Sanyal

    Advances in Neural Information Processing Systems (NeurIPS) 2025; Theory and Practice of Differential Privacy (TPDP) 2025

    arXiv TPDP version