About the Conference. In recent years, the combination of deep learning techniques and numerical methods has gained increasing interest as a research area in applied mathematics, especially in the approximation of PDEs. One promising paradigm that has emerged is operator learning. These innovative methodologies are crucial since traditional numerical methods face significant challenges with complex, high-dimensional PDEs that make them…

Adaptive finite element interpolated neural networks   We propose a general framework for solving forward and inverse problems constrained by partial differential equations, where we interpolate neural networks onto adaptive finite element spaces to represent the (partial) unknowns, and define loss functions with dual norms of the finite element residual of the interpolation. The framework…

First announcement for the conference: “From Kinetic Theory to Data Science and Related Topics” Università di Pavia, Aula Foscolo Pavia, Italy December 6-7, 2022     The detailed program and the registration for the workshop are available on the website:   https://ktds2022.wordpress.com/   Please have a look at the webpage for additional information and forward…