FOUNDRY is bringing together six research teams across France to advance the theoretical foundations of robustness and reliability in AI. Spanning Grenoble, Lyon, Paris, and Lille, the consortium forms a cohesive national network that combines expertise in machine learning, optimization, game theory, statistics, and information theory.
GHOST
GHOST is a joint CNRS/Inria team based at the Laboratoire d’Informatique de Grenoble, and its activities in FOUNDRY are headed by Panayotis Mertikopoulos. It is bringing together expertise on game theory, optimization, and reinforcement learning, with a particular focus on multi-agent learning, Markov decision processes, mechanism design, and adaptive optimization.
Faculty
- Panayotis Mertikopoulos (project coordinator)
- Jonatha Anselmi (Inria CR)
- Dorian Baudry (Inria CR)
- Mathieu Besançon (Inria CR)
- Bruno Gaujal (Inria DR)
- Bary Pradelski (CNRS CR; currently at Oxford)
Post-docs
PhD students
- Rim Alhajal
- Waïss Azizian
- Pierre-Louis Cauvin
- Karl Gottlieb
- Davide Legacci
ENS Lyon
Based in Lyon, the ENSL team is headed by Aurélien Garivier, and it focuses on stochastic modeling, sequential statistics, and reinforcement learning. In recent years, the team’s most impactful contributions include an original approach to the study of the complexity of noisy optimization and other sequential decision problems as well as (multi-)quantile estimation.
Faculty
- Aurélien Garivier (Professor, ENSL)
Post-docs
PhD students
FAIRPLAY
FAIRPLAY is a joint Inria-Criteo-ENSAE team based in Paris, and led by Patrick Loiseau. It aims to develop and analyze algorithms that learn alongside users, focusing in particular on creating procedures capable of performing traditional learning tasks—such as prediction, decision-making, and explanation—when the data is generated or supplied by strategic agents or in the presence of competing learning agents. The team’s work emphasizes societal and ethical considerations, particularly regarding the fairness and privacy of the users involved.
Faculty
Post-docs
PhD students
LTCI
LTCI is the Information and Communications Laboratory of Télécom Paris. It is based in Paris, and headed by Florence d’Alché-Buc. It focuses on statistical and trustworthy machine learning, with expertise in Monte Carlo methods, extreme value theory, and learning from complex and structured data.
Faculty
Post-docs
PhD students
MILES
The MILES team (Machine Intelligence and Learning Systems) is a joint research team shosted by the LAMSADE research unit at Université Paris Dauphine-PSL, and headed by Yann Chevaleyre. It focuses on adversarial machine learning models, game theory, and computational social choice, and it aim at making significant advances in the field of machine learning and data science, in terms of both theoretical foundations and concrete applications.
SCOOL
SCOOL is an Inria project-team based in Lille, and headed by Émilie Kaufmann. The team’s main research topic is the study of sequential decision making under uncertainty, and most of its activities are related to either bandit problems, or reinforcement learning. Through collaborations, the team is working on their application in various fields, from medicine and agriculture, to ecology and sustainable development.
Affiliates
- Franck Iutzeler (Professor, Institut Mathématique de Toulouse)
- Jean-Michel Loubès (Professor, Institut Mathématique de Toulouse)
- Jérôme Malick (CNRS DR, Laboratoire Jean Kuntzmann)