An r -adaptive finite element method using neural networks for parametric self-adjoint elliptic problems
Year: 2025
Type: article
Abstract: This work proposes an r -adaptive finite element method (FEM) using neural networks (NNs). The method employs the Ritz energy functional as the loss function, currently limiting its applicability to symmetric and coercive problems, such as those arising from self-adjoint elliptic problems. The objective of the NN optimization is to determine the mesh node locations. For simplicity, these locations are assumed to form a tensor product structure in higher dimensions. The method is designed to solve parametric partial differential equations (PDEs). The resulting parametric r -adapted mesh generated by the NN is solved for each PDE parameter instance with a standard FEM. Consequently, the proposed approach retains the robustness and reliability guarantees of the FEM for each parameter instance, while the NN optimization adaptively adjusts the mesh node locations. The construction of FEM matrices and load vectors is implemented such that their derivatives with respect to mesh node locations, required for NN training, can be efficiently computed using automatic differentiation. However, the linear equation solver does not need to be differentiable, enabling the use of efficient, readily available ‘out-of-the-box’ solvers. The method’s performance is demonstrated on parametric one- and two-dimensional Poisson problems.
Source: Journal of Computational Physics
Language: English
Subfield: Statistical and Nonlinear Physics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,980
Funders: European Commission, Eusko Jaurlaritza, Euskal Herriko Unibertsitatea, Basque Center for Applied Mathematics, Agencia Nacional de Investigación y Desarrollo, HORIZON EUROPE Framework Programme, European Regional Development Fund, Agencia Estatal de Investigación, Office of Naval Research, Fondo Nacional de Desarrollo Científico y Tecnológico, H2020 Marie Skłodowska-Curie Actions, HORIZON EUROPE Marie Sklodowska-Curie Actions, NextGenerationEU
Awards: G1278513833, G1612158077, G1633674249, G2262748287, G2785478817, G3170287466, G318412399, G3480869486, G3607574089, G3681454997, G403708639, G4042783231, G4073569113, G4241082588, G425668148, G4993905653, G5049803760, G5140928046, G5195188938, G542055026, G5451035857, G5517553624, G55232499, G5643777105, G5852410385, G5943232429, G6072587662, FINITE ELEMENT METHODS FOR HAMILTONIAN PARTIAL DIFFERENTIAL EQUATIONS AND TRANSMISSION PROBLEMS, G6294024929, G6437700672, G6481147803, G6520718613, G6882141640, G7084143925, G7564622933, DEEP (PHYSICS-INFORMED) COLLABORATION NETWORK, G7713090867, G7727141062, G7748812462, G7829631415, NATIONAL CENTER FOR ARTIFICIAL INTELLIGENCE RESEARCH (NAIR), G8939903727
OpenAlex ID: W4415313352
