A Localized Hybrid Finite Element--Random Feature Method for Elliptic Interface Problems on Unfitted Meshes
Abstract
We develop a localized hybrid finite element--random feature method (FEM--RFM) for elliptic interface problems on unfitted background meshes. The central idea is to use the random feature method (RFM) only in the narrow band formed by interface-cut cells of a uniform Cartesian background mesh,while retaining standard Qp finite elements on the remaining uncut cells. Within the band, local RFM representations are used on the two physical sides of the interface. The prescribed jumps in the solution and normal flux are incorporated through residual equations involving the two RFM representations on the physical interface Γ, while the finite element method (FEM) and RFM components are coupled across a separate grid-aligned artificial interface Γc through trace and variational normal-flux residuals. A Schur reduction eliminates the interior finite element unknowns through sparse solves while retaining their Galerkin equations as equality constraints. The remaining band and interface residuals are minimized in a reduced least-squares problem involving only the coupling nodal values and random-feature coefficients. The method is assessed numerically on a range of two- and three-dimensional interface problems involving high coefficient contrasts, nonsmooth and multiple interfaces, and irregular outer boundaries, together with an application-oriented thermal-homogenization problem for a periodic bicontinuous composite.
Related
- Geometry-Conforming Finite Element Methods for Interface Problems on Fitted and Unfitted Meshes
- Robust preconditioners for finite element approximations of control-state constrained optimal control problems governed by elliptic equations
- Error Analysis for Solving Elliptic Interface Problems with Randomized Neural Networks
- A Hybrid Iterative Deep Ritz Method for Elliptic Interface Problems
- A Domain-specific adaptation network for elliptic interface problems
BibTeX
@misc{lang2026localized,
title = {{A Localized Hybrid Finite Element--Random Feature Method for Elliptic Interface Problems on Unfitted Meshes}},
author = {Siyuan Lang and Zhiyue Zhang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.11688},
eprint = {2610.11688},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.11688},
}
RIS
TY - UNPB
TI - A Localized Hybrid Finite Element--Random Feature Method for Elliptic Interface Problems on Unfitted Meshes
AU - Siyuan Lang
AU - Zhiyue Zhang
PY - 2026
DA - 2026/10/08
PB - arXiv
DO - 10.48550/arxiv.2610.11688
UR - https://arxiv.org/abs/2610.11688
AB - We develop a localized hybrid finite element--random feature method (FEM--RFM) for elliptic interface problems on unfitted background meshes. The central idea is to use the random feature method (RFM) only in the narrow band formed by interface-cut cells of a uniform Cartesian background mesh,while retaining standard Q_p finite elements on the remaining uncut cells. Within the band, local RFM representations are used on the two physical sides of the interface. The prescribed jumps in the solution and normal flux are incorporated through residual equations involving the two RFM representations on the physical interface Γ, while the finite element method (FEM) and RFM components are coupled across a separate grid-aligned artificial interface Γ_c through trace and variational normal-flux residuals. A Schur reduction eliminates the interior finite element unknowns through sparse solves while retaining their Galerkin equations as equality constraints. The remaining band and interface residuals are minimized in a reduced least-squares problem involving only the coupling nodal values and random-feature coefficients. The method is assessed numerically on a range of two- and three-dimensional interface problems involving high coefficient contrasts, nonsmooth and multiple interfaces, and irregular outer boundaries, together with an application-oriented thermal-homogenization problem for a periodic bicontinuous composite.
ER -