معرفی
Damien Garreau is Professor for the Theory of Machine Learning at Julius-Maximilians-Universität Würzburg, Germany. Until March 2024 he served as Associate Professor in the Probability and Statistics team of the J. A. Dieudonné laboratory at Université Côte d'Azur and was a member of the Inria Maasai team in Sophia-Antipolis. Earlier positions include post-doctoral research at the Max Planck Institute for Intelligent Systems in Tübingen and PhD studies in the Inria Sierra team in Paris.
Education & Career Path
- PhD, Inria Sierra team, Paris – advisors Sylvain Arlot & Gérard Biau
- Post-doc, Max Planck Institute for Intelligent Systems, Tübingen – mentor Ulrike von Luxburg
- Associate Professor, Université Côte d’Azur / Inria Maasai (until March 2024)
- Professor for Theory of Machine Learning, Julius-Maximilians-Universität Würzburg (since 2024)
Research Focus
Garreau’s research centers on trustworthy machine learning. He investigates how to explain, audit, and robustify modern AI systems, with particular emphasis on post-hoc interpretability, statistical guarantees of explanation methods, fairness, and causality. Representative contributions include theoretical analyses of LIME and Anchors, novel explanation methods such as SMACE and GLEAMS, and practical tools for vision and NLP that remain faithful under adversarial or out-of-distribution settings.
Across computer vision, natural-language processing, and healthcare applications, his work bridges rigorous theory with impactful algorithms, advancing the societal goal of deploying AI systems whose decisions can be trusted and understood by humans.
Scientific Awards & Recognition
- Best Paper Award, ECML 2024
- Area Chair, ICML 2025
- ANR JCJC Grant NIM-ML (2021–2025)
- Université franco-allemande support for Winter School on Causality and Explainable AI
Advising, Grants & Collaborative Projects
Garreau has successfully supervised or co-supervised a growing cohort of doctoral and master’s students, including Gianluigi Lopardo, Kensuke Mitsuzawa, Martin Charachon, Jonas Wacker, Samuel, Antonio, Magamed, Arthur Assad, Charbel Yahchouchi, and Mariana Chaves. He is the PI of the ANR JCJC project NIM-ML, whose goal is to develop next-generation interpretability methods endowed with statistical guarantees. He co-organizes the annual Winter School on Causality and Explainable AI, fostering Franco-German academic exchange.
Labs & Teams
Since 2024 he leads the Professorship for the Theory of Machine Learning at Julius-Maximilians-Universität Würzburg. Previously he was a core member of the Maasai Inria team on the Sophia-Antipolis campus, and an active collaborator of the J. A. Dieudonné mathematics laboratory. He maintains strong ties with the TML group at the Max Planck Institute for Intelligent Systems and regularly hosts joint visitors and workshops.
