The integration of physical knowledge into learning models is a fast-growing theme aimed at improving the modeling of complex physical phenomena covering a wide spectrum of applications. The Physics-Aware Machine Learning working group is organizing a one-day event to bring together researchers working in this field with a data-driven approach.
The day will focus on foundation models and large-scale models of physics, with several invited talks on the subject in the morning. The afternoon will be devoted to lectures, posters with short presentations and time for exchanges between participants.
Page of the event on the GdR IASIS website (in french): https://gdr-iasis.cnrs.fr/reunions/journee-apprentissage-informe-par-la-physique-et-grands-modeles/
Researchers are invited to present their work can submit their proposal (title and extended abstract of no more than 2 pages) on openreview via this link: https://openreview.net/group?id=cnrs.fr/GdR-IASIS/2025/WPAML. Contributions may cover, but are not limited to, the following topics: foundation models, datasets, generative models, applications to various fields of physics (computational fluid dynamics, weather forecasting, materials, surfaces, cosmology, computer graphics, etc.). Deadline: May 18.
Johannes Bransdetter (Johannes Kepler University Linz)
Guillaume Couairon (Inria Paris)
Laure Raynaud (CNRM Toulouse)
Date: Tuesday 10 June 2025
Location: Amphi Astier, bâtiment Esclangon, RDC, Sorbonne université, Campus Pierre et Marie Curie, 4 place Jussieu 75005 Paris