👀

Indeed, could not be more timely! Excited to share the joys of statistical physics of learning with the crowd at Nordita! :)

On the Geometry of Regularization in Adversarial Training: High-Dimensional Asymptotics and Generalization Bounds ift.tt/dup5fOC

Interested in coming for a postdoc at Ecole Normale Supérieure in Paris? To work in statistical and theoretical physics? With a very prestigious position? Check these offers and apply! academicjobsonline.org/ajo/jobs/28737 academicjobsonline.org/ajo/jobs/27869

For adversarial training, the choice of regularization norm matters - a lot! Choosing the *dual* norm of the perturbation (often) helps! See arxiv.org/abs/2410.16073 for exact bounds and theory. With Matteo Vilucchio, Nikos Tsilivis & Bruno Loureiro, NYU Center for Data Science École normale supérieure | PSL EPFL

🚨Only two days left for registering to the Les Houches workshop "Towards a theory for typical-case algorithmic hardness"! ❄️🏔️ 🌐leshouches-algorithms.github.io

The long-awaited collection of lecture notes from the Summer School on Statistical Physics & Machine Learning, Les Houches 2022, is now published in JStatMech iopscience.iop.org/collections/js… . I am particularly proud of the works the school inspired; see section 3 of the editorial.

"My definition of physics is that physics is not what you’re working on, but how you’re working on it. If you have the attitude of someone who comes from physics, it’s a physics problem." nature.com/articles/d4158…

Our Nordita Winter School on Machine Learning & Physics has officially kicked off yesterday! Huge thanks to our amazing lecturers Bruno Loureiro Arno Solin for sharing their expertise and inspiring the next generation of scientists! The learning journey is in full swing 📈

Presentation on Thursday, August 7, given by Bruno Loureiro Bruno Loureiro, École normale supérieure | PSL Paris, on “An overview on Gaussian Multi-Index Model" during the conference organized by Krzakala Florent & Lenka Zdeborova EPFL Université Côte d'Azur Università di Corsica CNRS Provence et Corse #statisticalPhysics #machinelearning