Track / Overview

The exchange of ideas between the fields of Artificial Intelligence (AI) and Physics share a long and rich history that led to major developments in both sides. The recent breakthroughs in AI and the rapid growth of Machine Learning (ML) based technology in our day-a-day life has fostered a fresh wave of interest in the synergies between these fields. On one hand, the technological progress in AI has allowed for the fruitful application of Machine Learning techniques in theoretical modelling and experimental data analysis. Indeed, from Molecular Modelling to String Theory, it is hard to name one subfield of Physics which has not been impacted by Machine Learning in the past decade. On the other hand, the lack of a unified theoretical framework allowing to make sense of the challenges raised by the ever-growing practice of Machine Learning has drawn the interest of theoretical Physicists to this field. From the historical connections with Statistical Physics to new inputs coming from High-Energy Physics, ideas and tools from Physics had a major impact in the recent progress of Machine Learning theory.

This track will investigate applications of ML to Physics, ranging from particle physics to biophysics and astronomy, and advancements in understanding ML using tools from Statistical Physics.

Track / Schedule

Introduction

Renormalization Group Theory and Machine Learning

With Giulio Biroli

Statistical Physics Insights on Stochastic Gradient Descent

With Francesca Mignacco

Universal mean field upper bound for the generalization gap of Deep Neural Networks

With Pietro Rotondo

AI Pontryagin or: How Artificial Neural Networks Learn to Control Dynamical Systems

With Thomas Asikis

Break & Posters

Stable Convolutional Neural Networks for Biomedical Image Reconstruction

With Michael Unser

Automating Myself Away? Discovering PDEs from Data Using Neural Networks

With Gert-Jan Both

Atmospheric Physics-Guided Machine Learning

With Tom Beucler

Meta-Learning for Fast Simulation of Multiple Calorimeter Responses

With Dalila Salamani

Track / Speakers

Francesca Mignacco

Ph.D. student, Institute of Theoretical Physics, Paris-Saclay

Gert-Jan Both

Researcher, Pasqal

Michael Unser

Professor, EPFL

Thomas Asikis

Dr, ETH

Dalila Salamani

CERN

Pietro Rotondo

INFN Fellini Fellow

Tom Beucler

Assistant Professor, UNIL

Giulio Biroli

Professor, ENS

Track / Co-organizers

Bruno Loureiro

Scientist, EPFL

Jonathan Dong

Scientist, EPFL

Vittorio Erba

Scientist, EPFL

AMLD EPFL 2022 / Tracks & talks

AMLD Keynote Session – Monday morning

Marcel Salathé, Lenka Zdeborová, Carmela Troncoso, Chiara Enderle, Patrick Barbey, Thomas Wolf, Gunther Jansen, Laure Willemin, Simon Hefti, Arthur Gassner

10:00-12:00 March 28Auditorium A

AI & Physics

Francesca Mignacco, Gert-Jan Both, Michael Unser, Thomas Asikis, Dalila Salamani, Pietro Rotondo, Tom Beucler, Giulio Biroli

12:30-18:00 March 285BC

AI & Pharma

Asif Jan, Jonas Richiardi, Patrick Schwab, Naghmeh Ghazaleh, Alexander Büsser, Carlos Ciller, Caibin Sheng, Silvia Zaoli, Félix Balazard, Giulia Capestro, Marianna Rapsomaniki, Martijn van Attekum

13:30-17:30 March 281BC

AMLD / Global partners