!! Télécharger l'affiche et le programme (PDF): ici 

Jour 2

le 28 Mai 2026

Matinée de formation

 8h45 à 9h : Accueil des participants

 9h à 12h30 (avec une pause) : LLM & L'IA agéntique

Intervenant : Maria Manera (Post-Doc, l'Université de Torino))

12h30 à 14h : Déjeuner 

Après-midi de présentation de travaux

14h à 15h  :  Keynote invité Vianney Perchet (CREST, ENSAE, Criteo Ai Lab)

Learning in matching markets

In this presentation, I will consider the problem of learning the optimum (stable) matching in a two-sided market. We will first discuss the case of preferences with ties (which happen quite naturally in practical applications where grades/stars are given by users to commodities); this case is often overlooked by the literature, yet learning is quite intricate. I will explain why and how it can still happen. The second part of the presentation will focus on the case where matching are constructed sequentially, with prior - but noisy - information on the preferences. Here again, learning is intricate, but achievable.

This is based on two joint papers with S. Lin, S. Mauras and N. Merlis.

15h à 15h40  : Presentation 4 previsionnelle : Moncif El-Mouden (Université Paris Est Créteil, LACL et ERUDITE)

 From Legal Text to Quantitative Insights: Agentic AI for Competition Authority Decisions

15h40 à 16h  :  Pause

16h à 16h40  :  Presentation 5 previsionnelle :  Olha NAHORNA (Bordeaux School of Economics, CNRS)

Agent evaluation

16h40 à 17h20  :  Presentation 6 previsionnelle : Hammou El-Otmany (Toulouse School of Economics, University of Toulouse Capitole, France, Université de Pau et des Pays de l'Adour [UPPA])

Explainable DiD–Double Machine Learning for Housing Price Prediction: Evidence from France's 2021 Energy-Label Reform

 Rejoignez-nous en visioconférence en accès libre (uniquement l'après-midi) !