Bruno Loureiro

I am a CNRS researcher in the DATA team, part of the Centre for Data Science at the École Normale Supérieure in Paris. My work lies at the intersection of machine learning and statistical mechanics. I also serve as a Professeur Attaché at Université Paris Sciences et Lettres (PSL), where I teach in both undergraduate and graduate programs across its affiliated institutions.

If you are interested in my research, you can find a complete list of my publications here.

news

Jan 26, 2026	Our works: Scaling Laws and Spectra of Shallow Neural Networks in the Feature Learning Regime Fast Escape, Slow Convergence: Learning Dynamics of Phase Retrieval under Power-Law Data Statistical Advantage of Softmax Attention: Insights from Single-Location Regression were accepted for ICLR 2026. Come see our posters in Rio de Janeiro!
Jan 09, 2026	I have successfully defended my Habilitation à diriger des recherches (HDR) “Statistical Physics of two-layer neural networks: from kernels to feature learning” in Computer Science ! You can check the manuscript out here.
Dec 09, 2025	The PEPR IA project “Mathematics of Generation: an Interdisciplinary Cross Analysis of Loss Landscapes” (MAGICALL), which I am a co-PI in collaboration with Giulio Biroli (LPENS), Marylou Gabrié (LPENS) and Anna Korba (ENSAE) has been selected !
Oct 09, 2025	The recordings of the lectures I gave at the INFORMS Applied Probability Society summer school are now available on this YouTube channel.
Sep 18, 2025	Our works: Optimal Spectral Transitions in High-Dimensional Multi-Index Models Asymptotics of SGD in Sequence-Single Index Models and Single-Layer Attention Networks were accepted for NeurIPS 2025. Come see our posters in San Diego!