Recurrent Neural Networks as Optimal Mesh Refinement Strategies

DOI10.1016/J.CAMWA.2021.05.018arXiv1909.04275MaRDI QIDQ6324992FDOQ6324992

Publication date: 10 September 2019

Abstract: We show that an optimal finite element mesh refinement algorithm for a prototypical elliptic PDE can be learned by a recurrent neural network with a fixed number of trainable parameters independent of the desired accuracy and the input size, i.e., number of elements of the mesh. Moreover, for a general class of PDEs with solutions which are well-approximated by deep neural networks, we show that an optimal mesh refinement strategy can be learned by recurrent neural networks. This includes problems for which no optimal adaptive strategy is known yet.

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Boundary value problems for second-order elliptic equations (35J25) Stability and convergence of numerical methods for boundary value problems involving PDEs (65N12) Error bounds for boundary value problems involving PDEs (65N15) Finite element, Rayleigh-Ritz and Galerkin methods for boundary value problems involving PDEs (65N30) Mesh generation, refinement, and adaptive methods for boundary value problems involving PDEs (65N50) PDEs on graphs and networks (ramified or polygonal spaces) (35R02)

This page was built for publication: Recurrent Neural Networks as Optimal Mesh Refinement Strategies

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6324992)