Dear All,

You are most cordially invited to the ICTP Colloquium "Generative AI and Diffusion Models: A Statistical Physics Perspective" by Giulio Biroli from École Normale Supérieure, Paris.

The Colloquium will take place on Wednesday 9 December 2026 in the Budinich Lecture Hall at ICTP at 2:00PM CET.  Light refreshments will be served at end of the event in the Leonardo Building entrance hall.

Generative AI represents a groundbreaking development within the broader “Machine Learning Revolution,” significantly influencing technology, science, and society. In this talk, Prof. Giulio Biroli will focus on remarkable connections between generative AI, in particular diffusion models, and statistical physics. Diffusion models are the state of the art 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. These diffusion models start from a simple white noise input and make it evolve through a Langevin process to generate complex outputs such as images, videos, and sounds. Prof. Biroli will show that statistical physics provides principles and methods to characterise this generation process. The emergence of features can be understood through the lens of symmetry breaking and dynamical phase transitions, whereas the emergence of creativity - or its absence (memorisation) - can be explained using  theories of disordered systems, and dynamics in complex energy landscapes.  Prof. Biroli will conclude by discussing the fundamental reasons behind the success of diffusion models compared to earlier generative methods, highlighting a deep connection between their multi-scale generation mechanism and the renormalisation group analysis of physical systems. 

About the speaker
Giulio Biroli is professor of theoretical physics at the École Normale Supérieure in Paris.
From 2002 to 2018 he held a researcher position at the Institute for Theoretical Physics (CEA) Giulio Biroli has a long-standing interest in slow dynamics of classical and quantum systems, in disordered and glassy systems and in inter-disciplinary applications of statistical physics ranging from computer science, probability theory, and theoretical ecology. In recent years, part of his research has been devoted to the theory of machine learning leveraging ideas and methods rooted in statistical physics.
 
The event will be livetreamed at:
http://ictp.it/livestream

For info, please check the following link: https://indico.ictp.it/event/11552/

Looking forward to your participation.

With best regards, 

Director's Office, ICTP