Fiche de DE BEZENAC Emmanuel

Titre : Modeling Physical Processes with Deep Learning: A Dynamical Systems Approach
Date : 21 octobre 2021
Heure: 15:00
Résumé : Deep Learning has emerged as a predominant tool for AI, and has already abundant applications in fields where data is abundant and access to prior knowledge is difficult. This is not necessarily the case for natural sciences, and in particular, for physical processes and dynamical systems. These have been the object of study since centuries, and knowledge as well as algorithms and methods have been developped. This thesis has two main objectives. The first considers the role that deep learning has to play in this vast ecosystem of theory and tools. We will attempt to answer this general question through the problem of modelling complex physical processes where the data can be leveraged in order to make up for lacking prior knowledge. The second objective focuses on how perspectives, insights and tools from the field of study of physical processes and dynamical systems can be applied in the context of deep learning, in order to gain a better understanding and develop algorithms.
Lieu : Sorbonne Université