Nicolas Schreuder
About me
I am a CNRS research scientist (“chargé de recherche”), based at the Laboratoire d'Informatique Gaspard-Monge.
Before joining CNRS, I was a postdoctoral researcher working with Lorenzo Rosasco at Università di Genova, Italy.
I hold a PhD in Statistics from CREST (Center for Research in Economics and Statistics) and Institut Polytechnique de Paris, where I was supervised by Arnak Dalalyan and Victor-Emmanuel Brunel. My PhD dissertation is available here.
I have primarily worked on fairness in machine learning from a statistical learning perspective. More broadly, I am interested in designing and analyzing learning algorithms that incorporate key aspects such as fairness, robustness, and privacy.
Preprints
- Sholom Schechtman, NS (2025). The late-stage training dynamics of (stochastic) subgradient descent on homogeneous neural networks. [arXiv:2502.05668]
- Antoine Chatalic, Marco Letizia, NS, Lorenzo Rosasco (2025). An Efficient Permutation-Based Kernel Two-Sample Test. [arXiv:2502.13570]
- Antoine Chatalic, NS, Ernesto De Vito, Lorenzo Rosasco (2023). Efficient Numerical Integration in Reproducing Kernel Hilbert Spaces via Leverage Scores Sampling. [arXiv:2311.13548]
Publications
- Ziyad Benomar, Evgenii Chzhen, NS, Vianney Perchet (2024). Addressing bias in online selection with limited budget of comparisons. Accepted at NeurIPS 2024. [arXiv:2303.09205]
- Solenne Gaucher, NS, Evgenii Chzhen (2023). Fair learning with Wasserstein barycenters for non-decomposable performance measures. AISTATS 2023. [PMLR 206:2436-2459]
- Evgenii Chzhen, NS (2022). A minimax framework for quantifying risk-fairness trade-off in regression. Annals of Statistics. [10.1214/22-AOS2198]
- Antoine Chatalic, NS, Alessandro Rudi, Lorenzo Rosasco (2022). Nyström Kernel Mean Embeddings. ICML 2022. [PMLR 162:3006-3024]
- NS (2021). A study of some trade-offs in statistical learning: online learning, generative models and fairness. PhD manuscript. Institut Polytechnique de Paris. [tel-03435618]
- NS, Evgenii Chzhen (2021). Classification with abstention but without disparities. UAI 2021. Runner-up for best student paper award. [PMLR 161:1227-1236]
- NS, Victor-Emmanuel Brunel, Arnak Dalalyan (2021). Statistical guarantees for generative models without domination. ALT 2021. [PMLR 132:1051-1071]
- Evgenii Chzhen, NS (2020). An example of prediction which complies with Demographic Parity and equalizes group-wise risks in the context of regression. NeurIPS 2020 Workshop on Algorithmic Fairness through the Lens of Causality and Interpretability. [arXiv:2011.07158]
- NS (2020). Bounding the expectation of the supremum of empirical processes indexed by Hölder classes. Math. Meth. Stat. 29, 76–86 (2020). [link]
- NS, Victor-Emmanuel Brunel, Arnak Dalalyan (2019). A nonasymptotic law of iterated logarithm for general M-estimators. AISTATS 2020. [PMLR 108:1331-1341]
Education
- 2018 - 2021: PhD candidate, CREST, ENSAE Paris, IPP.
- 2017 - 2018: Master’s in Mathematics, Vision, and Learning (MVA), ENS Paris-Saclay.
- 2015 - 2018: Master’s-level Engineering Degree in Statistics, ENSAE Paris.
- 2012 - 2015: BSc in Applied Mathematics, Université Paris Dauphine-PSL.
Teaching
Professeur attaché at PSL (2023-...)
- Statistical learning, Double Bachelor's degree in AI and Organizational Sciences, Université Paris-Dauphine.
- Deep learning, "Training for Academics" program, PSL.
Adjunct professor at IPP (2023-...)
- Introduction to ML, MScT Data and Economics for Public Policy.
- Fairness in ML, ENSAE Master’s-level Engineering Degree program.
Teaching assistant at ENSAE Paris (2019-2021)
- Theoretical foundations of ML and advanced ML with V. Perchet.
- Measure theory with A. Korba.
- Applied statistics project with J. Depersin.
- Numerical analysis and numerical linear algebra with S. M. Kaber.