Julien BENISTANT – Computational mechanisms of learning in social networks

Julien BENISTANT

Depuis septembre 2023, je suis maître de conférences en économie à l’Université de Lille. Auparavant, j’ai effectué un post doctorat à l’Institut des Sciences Cognitives à Lyon dans l’équipe neuro-économie. Enfin, j’ai soutenu une thèse en économie en 2019 à l’Université Lyon 2 en économie comportementale et expérimentale.

Résumé

Learning and decision making is not done by isolated individuals. It is also shaped by information aggregated from our social networks. In this project we test different models accounting for the computational mechanisms reflecting this aggregation process. We confront two family of model, models adapted from the canonical DeGroot model and models derived from the reinforcement learning framework. Using data from a published experiment, we show that, despite its simplicity, a model derived from the DeGroot model family best explain how individuals learn from their social network. We currently are extending this work by testing the different models onto others datasets.

Séminaire ouvert à tous, sur inscription auprès de Marie Pelé (marie.pele@univ-catholille.fr).