Theresa Stadler

Postdoctoral Researcher & Lecturer at EPFL.

PortraitCropped.JPG

I am a Postdoctoral Researcher & Lecturer at the Security & Privacy Engineering Lab at EPFL (Switzerland) led by Carmela Troncoso.

My research focuses on the potentially harmful aspects of data processing systems centred around questions such as: How to evaluate the privacy properties of opaque data processing systems? What are the privacy limits of machine learning-based applications? What learning tasks are solvable under good privacy and good utility simultaneously? My work has been featured on multiple national media outlets and continues to inform policy makers on a national and European level.

Before joining EPFL, I previously worked as a researcher for Privitar, a London-based start-up, where I developed enterprise software that implements privacy-enhancing technologies and aims to makes these technologies available to organisations at scale. I hold a PhD in Computer Science from EPFL (Switzerland) and a Master’s degree in Computational Neuroscience (Biomathematics) from the University of Tübingen (Germany).

news

Jun 02, 2025 Interview with Svea Eckert and Eva Wolfange on the “They talk Tech” podcast and for heise online about privacy engineering.
May 19, 2025 Invited talk at the Responsible AI Seminar at Nokia Bell Labs.
Nov 01, 2024 Public defence of my PhD thesis On the Fundamental Limits of Privacy Enhancing Technologies
Sep 10, 2024 Talk and panel discussion at the Synthetic Data for AI Conference organised by the European Commission.
Aug 25, 2024 Lecture at the Brussels Privacy Hub Global Summer Academy for Privacy Law 2024 on Privacy-preserving Data Sharing in the Context of the European Data Strategy

latest posts

selected publications

  1. Conference
    The Fundamental Limits of Least-Privilege Learning
    Theresa Stadler, Bogdan Kulynych, Nicoals Papernot, and 2 more authors
    In Proceedings of the 41th International Conference on Machine Learning (ICML 24), 2024
  2. Conference
    Synthetic Data – Anonymisation Groundhog Day
    Theresa Stadler, Bristena Oprisanu, and Carmela Troncoso
    In 31st USENIX Security Symposium (USENIX Security 22), 2022
  3. Preprint
    Decentralized privacy-preserving proximity tracing
    Carmela Troncoso, Mathias Payer, Jean-Pierre Hubaux, and 8 more authors
    arXiv preprint arXiv:2005.12273, 2020