Principal-Part Decomposition for Neural Operator Learning of Dirichlet-to-Neumann Maps

DOI10.48550/arXiv.2606.25952MaRDI QIDQ7270043arXivFDO

Authors Wenjun Ying, Han Zhou, Shuo Ling

Publication date 24 June 2026

Published in arXiv

Tldr We propose Principal-Part Decomposed Neural Operators (PPDNO), a hybrid model that uses the boundary integral representation of Dirichlet-to-Neumann maps to decompose their principal behavior from geometry-dependent correction, approximating only the latter with a low-rank neural network.

Full work available at URL https://doi.org/10.48550/arXiv.2606.25952

arXiv classification

math.NA

This page was built for publication: Principal-Part Decomposition for Neural Operator Learning of Dirichlet-to-Neumann Maps

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7270043)