Lenka Zdeborová: Understanding machine learning via exactly solvable models

0
378

Prof. Lenka Zdeborova gave a lecture at CIIRC on 29.9.2021.

The lecture was organized by the IMPACT Project (http://impact.ciirc.cvut.cz) and ELLIS Unit Prague (http://www.ellisprague.eu/).

Abstract: Inspired by physics, where simple models are at the core of our theoretical understanding of the physical world, we study simple models of neural networks to clarify some of the open questions surrounding learning with neural networks. In this talk, I will describe some of our recent progress in this direction.

Bio: Lenka Zdeborová received a PhD in physics from University Paris-Sud and from Charles University in Prague in 2008. She then spent two years in the Los Alamos National Laboratory as the Director’s Postdoctoral Fellow. She was then Director of Research at the Institut de physique théorique du CEA Saclay, Gif-sur-Yvette, France, During the second half of 2020 she was Visiting Scientist at the Simons Institute, Berkeley. Since September 2020 she is an Associate Professor of Physics and of Computer Science and Communication Systems in the Schools of Basic Sciences (SB) and Computer and Communications Sciences (IC) at EPFL. In 2014, Lenka was awarded the CNRS bronze medal, in 2016 the Philippe Meyer prize in theoretical physics and an ERC Starting Grant, in 2018 the Irène Joliot-Curie prize, in 2020 she delivered the AMS Josiah Willard Gibbs lecture. She is an editorial board member for Journal of Physics A, Physical review E, Physical Review X, SIMODS, and Information and Inference. Lenka’s expertise is in applications of methods developed in statistical physics, such as advanced mean field methods, replica method and related message passing algorithms, to problems in machine learning, signal processing, inference and optimization.