MENU

You are here

Learning solution operators for PDEs with uncertainty

Speaker: 
Emilia Magnani
Institution: 
University of Tubingen
Schedule: 
Friday, June 21, 2024 - 14:00
Abstract: 

Modeling dynamical systems, e.g. in climate and engineering sciences, often necessitates solving partial differential equations (PDEs). In this talk we consider neural operators, recent deep architectures that have shown promising results in tackling the task of learning solution operators for PDEs. The current state of the art for these models lacks explicit uncertainty quantification. Our approach offers a practical and theoretically sound way to apply the linearized Laplace approximation to neural operators to provide uncertainty estimates. Moreover, we introduce a novel framework for Bayesian uncertainty quantification in neural operators using the framework of function-valued Gaussian processes.

Sign in