Séminaire de Marylou Gabrié: Accelerating Scientific Computations with Learning: the case of adaptive Monte Carlo

  • Culture scientifique
Publié le 7 février 2022 – Mis à jour le 16 janvier 2023
Date(s)

le 25 mars 2022

11:00
Lieu(x)
Site Valrose et on-line
Gabrié
Gabrié

Accelerating Scientific Computations with Learning: the case of adaptive Monte Carlo

Marylou Gabrié, CMAP, Ecole Polytechnique

In many applications in computational sciences and statistical inference, one seeks to compute expectations on complex high-dimensional distributions or high-dimensional integrals. These problems are often plagued by multi-modality/metastability; slow relaxation between unconnected modes leads to slow convergence of estimators. In this talk, I will present a strategy to enhance sampling with a class of deep generative models called Normalizing Flows. We will see how blending physics knowledge and learning is the winning cocktail to a tackle increasingly complicated systems.