On May 23rd 2025 was launched a new Groupement de Recherche (GDR) on scientific computing named “Calcul: Paradigmes, Parallélisme, Performance, Précision” (C4P). SHARP members were strongly involved in the initiative.
A note by P. Caillon and A. Allauzen from the SHARP project published on The Conversation. Pourquoi l’IA générative consomme-t-elle tant d’énergie ? Paul Caillon, Université Paris Dauphine – PSL et Alexandre Allauzen, Université Paris Dauphine – PSL DeepSeek défraye la chronique en proposant un modèle dont les performances seraient comparables à…
Several talks & posters from SHARP researchers @PEPR IA Days, CentraleSupelec, March 2025.
SHARP was present at the AI Action Summit, held at Ecole Polytechnique in February 2025, with several SHARP members presenting posters about recent advances of the project.
A 2-day internal brainstorming workshop of the SHARP project was held at ENS de Lyon on January 24-25 2025. Each day started with an invited seminar open to all registered participants Anne-Laure Ligozat “How to evaluate the carbon footprint in AI projects” Guillaume Charpiat “Neural Network Growth for Frugal AI”…
The SHARP projetct met on Nov 22nd 2024 at the Paris Santé Campus. The day combined a tutorial on diffusion models by David Picard, an overview of foundation models for physical sciences by Patrick Gallinari, several spotlight presentations. Much of the afternoon was then dedicated to a brainstorming session to…
A project meeting was held at ISIR (Paris) on July 4th 2024. The day was organized around several spotlight presentations as well as a tutorial. Morning spotlights “Learning to Simplify Large Computer Vision Problems” [Loic Landrieu] “Segmenting Historical Maps with Modern Labels” [Loic Landrieu] “Understanding diffusion models” [David Picard] “Efficient…
The official kickoff meeting of SHARP with the ANR was held on April 25th 2024.
SHARP’s second project meeting took place at ISIR on March 13th 2024. The scientific program involved spotlight presentations on: Asynchronous decentralized optimization of machine learning models (Stéphane Rivaud) Alternatives to backpropagation for learning neural networks (Paul Caillon) Physics-informed Deep Learning (Patrick Gallinari) Sparsity and quantization in deep networks: challenges and…