Pantheon SEMPARIS Le serveur des séminaires parisiens Paris

Statut Confirmé
Série FORUM-ENS
Domaines cond-mat.stat-mech
Date Mercredi 27 Novembre 2024
Heure 12:45
Institut LPENS
Salle 3 rue dUlm College de France
Nom de l'orateur Biroli
Prenom de l'orateur Giulio
Addresse email de l'orateur
Institution de l'orateur ENS
Titre Generative AI & Diffusion Models: a Statistical Physics Analysis
Résumé Generative AI represents a groundbreaking development within the broader “Machine Learning Revolution,” significantly influencing technology, science, and society. In this talk, I will focus on the state-of-the-art “diffusion models,” which are currently used to generate images, videos, and sounds. They are very fascinating algorithms for physicists, as they are very much connected to concepts from stochastic thermodynamics, particularly time-reversed Langevin dynamics. Diffusion models initiate from a simple white noise input and evolve it through a Langevin process to generate complex outputs such as images, videos, and sounds. I will show that statistical physics provides principles and methods to characterise this generation process. Specifically, I will discuss how phenomena such as the transition from memorization to generalization and the emergence of features can be understood through the lens of symmetry breaking, phase transitions, and disordered systems.
Numéro de preprint arXiv
Commentaires
Fichiers attachés

Pour obtenir l' affiche de ce séminaire : [ Postscript | PDF ]

[ Annonces ]    [ Abonnements ]    [ Archive ]    [ Aide ]    [ ]
[ English version ]