arXiv Science⌕ Search

arXiv subjects

Christos Xenophontos

Publications and source records attributed to Christos Xenophontos.

14 recordsLinked to original sources
arXiv2026

A simple shallow neural network for emulating the solution to singularly perturbed problems

We consider (feed-forward) Neural Networks (NNs) for the emulation of the solution to singularly perturbed second order boundary value problems, with two small parameters. We describe a shallow NN which exploits available asymptotic expansions for the solution. These additive decompositions into smooth and layer components, allow for derivative estimates which are explicit in the order of differentiation as well as the singular perturbation parameter(s) \cite{melenk, Irene, SX}. Utilizing such decompositions, we propose a simple NN for emulating the solution to such problems using the $\tanh$ activation function together with different training objectives, such as residual or energy minimization. The key idea is to augment the approximation space with suitable exponential functions, similar to enriched spaces in finite element methods, e.g.~\cite{Kellogg}. Numerical examples in one and two dimensions, including a smooth non-tensor-product domain, illustrate the resulting parameter-robust behavior over the tested perturbation ranges.

math.NA↗
arXiv2024

Neural Networks for Singular Perturbations

We prove deep neural network (DNN for short) expressivity rate bounds for solution sets of a model class of singularly perturbed, elliptic two-point boundary value problems, in Sobolev norms, on the bounded interval $(-1,1)$. We assume that the given source term and reaction coefficient are analytic in $[-1,1]$. We establish expression rate bounds in Sobolev norms in terms of the NN size which are uniform with respect to the singular perturbation parameter for several classes of DNN architectures. In particular, ReLU NNs, spiking NNs, and $\tanh$- and sigmoid-activated NNs. The latter activations can represent ``exponential boundary layer solution features'' explicitly, in the last hidden layer of the DNN, i.e. in a shallow subnetwork, and afford improved robust expression rate bounds in terms of the NN size. We prove that all DNN architectures allow robust exponential solution expression in so-called `energy' as well as in `balanced' Sobolev norms, for analytic input data.

math.NA↗
arXiv2023

Analytic regularity for a singularly perturbed reaction-convection-diffusion boundary value problem with two small parameters

We consider a second order, two-point, singularly perturbed boundary value problem, of reaction-convection-diffusion type with two small parameters, and we obtain regularity results for its solution. First we establish classical differentiability bounds that are explicit in the order of differentiation and the singular perturbation parameters. Next, for small values of these parameters we show that the solution can be decomposed into a smooth part, boundary layers at the two endpoints and a negligible remainder. Derivative estimates are obtained for each component of the solution, which again are explicit in the differentiation order and the singular perturbation parameters.

math.AP↗
arXiv2023

Balanced norm estimates for $rp$-Finite Element Methods applied to singularly perturbed fourth order boundary value problems

We establish robust exponential convergence for $rp$-Finite Element Methods (FEMs) applied to fourth order singularly perturbed boundary value problems, in a \emph{balanced norm} which is stronger than the usual energy norm associated with the problem. As a corollary, we get robust exponential convergence in the maximum norm. $r p$ FEMs are simply $p$ FEMs with possible repositioning of the (fixed number of) nodes. This is done for a $C^1$ Galerkin FEM in 1-D, and a $C^0$ mixed FEM in 2-D over domains with smooth boundary. In both cases we utilize the \emph{Spectral Boundary Layer} mesh.

math.NA↗
arXiv2022

An $hp$ Weak Galerkin FEM for singularly perturbed problems

We present the analysis for an $hp$ weak Galerkin-FEM for singularly perturbed reaction-convection-diffusion problems in one-dimension. Under the analyticity of the data assumption, we establish robust exponential convergence, when the error is measured in the energy norm, as the degree $p$ of the approximating polynomials is increased. The Spectral Boundary Layer mesh is used, which is the minimal (layer adapted) mesh for such problems. Numerical examples illustrating the theory are also presented.

math.NA↗
arXiv2021

A formula for the $n^{th}$ derivative of the quotient of two functions

As the title suggests, we give a formula for the $n^{th}$ derivative of a quotient of two functions, analogous to Leibniz's formula for the product. This particular note has remained unpublished since 2007 (available only my website), however it has received several citations. As a result, it now appears on ArXiv for future reference.

math.GM↗
arXiv2021

On the finite element approximation of fourth order singularly perturbed eigenvalue problems

We consider fourth order singularly perturbed eigenvalue problems in one-dimension and the approximation of their solution by the $h$ version of the Finite Element Method (FEM). In particular, we use piecewise Hermite polynomials of degree $p\geq 3$ defined on an {\emph{exponentially graded}} mesh. We show that the method converges uniformly, with respect to the singular perturbation parameter, at the optimal rate when the error in the eigenvalues is measured in absolute value and the error in the eigenvectors is measured in the energy norm. We also illustrate our theoretical findings through numerical computations for the case $p=3$.

math.NA↗
arXiv2019

An $hp$ finite element method for a singularly perturbed reaction-convection-diffusion boundary value problem with two small parameters

We consider a second order singularly perturbed boundary value problem, of reaction-convection-diffusion type with two small parameters, and the approximation of its solution by the $hp$ version of the Finite Element Method on the so-called {\emph{Spectral Boundary Layer}} mesh. We show that the method converges uniformly, with respect to both singular perturbation parameters, at an exponential rate when the error is measured in the energy norm. Numerical examples are also presented, which illustrate our theoretical findings.

math.NA↗
arXiv2019

Isogeometric Analysis for singularly perturbed problems in 1-D: a numerical study

We perform numerical experiments on one-dimensional singularly perturbed problems of reaction-convection-diffusion type, using isogeometric analysis. In particular, we use a Galerkin formulation with B-splines as basis functions. The question we address is: how should the knots be chosen in order to get uniform, exponential convergence in the maximum norm? We provide specific guidelines on how to achieve precisely this, for three different singularly perturbed problems.

math.NA↗
arXiv2017

Improved Reference Sampling and Subtraction: A Technique for Reducing the Read Noise of Near-infrared Detector Systems

Near-infrared array detectors, like the \JWST NIRSpec's Teledyne's H2RGs, often provide reference pixels and a reference output. These are used to remove correlated noise. Improved Reference Sampling and Subtraction (\IRSSquare, pronounced "IRS-square") is a statistical technique for using this reference information optimally in a least squares sense. Compared to "traditional" H2RG readout, \IRSSquare uses a different clocking pattern to interleave many more reference pixels into the data than is otherwise possible. Compared to standard reference correction techniques, \IRSSquare subtracts the reference pixels and reference output using a statistically optimized set of frequency dependent weights. The benefits include somewhat lower noise variance and much less obvious correlated noise. NIRSpec's \IRSSquare images are cosmetically clean, with less $1/f$ banding than in traditional data from the same system. This article describes the \IRSSquare clocking pattern and presents the equations that are needed to use \IRSSquare in systems other than NIRSpec. For NIRSpec, applying these equations is already an option in the calibration pipeline. As an aid to instrument builders, we provide our prototype \IRSSquare calibration software and sample \JWST NIRSpec data. The same techniques are applicable to other detector systems, including those based on Teledyne's H4RG arrays. The H4RG's "interleaved reference pixel readout" mode is effectively one \IRSSquare pattern.

astro-ph.IM↗
arXiv2015

Robust exponential convergence of $hp$-FEM in balanced norms for singularly perturbed reaction-diffusion equations

The $hp$-version of the finite element method is applied to a singularly perturbed reaction-diffusion equation posed in one- and two-dimensional domains with analytic boundary. On suitably designed \emph{Spectral Boundary Layer meshes}, robust exponential convergence in a balanced norm is shown. This balanced norm is stronger than the energy norm in that the boundary layers are $O(1)$ uniformly in the singular perturbation parameter. Robust exponential convergence in the maximum norm is also established. The theoretical findings are illustrated with two numerical experiments.

math.NA↗
arXiv2012

Analytic regularity for a singularly perturbed system of reaction-diffusion equations with multiple scales: proofs

We consider a coupled system of two singularly perturbed reaction-diffusion equations, with two small parameters $0< ε\le μ\le 1$, each multiplying the highest derivative in the equations. The presence of these parameters causes the solution(s) to have \emph{boundary layers} which overlap and interact, based on the relative size of $ε$ and $% μ$. We construct full asymptotic expansions together with error bounds that cover the complete range $0 < ε\leq μ\leq 1$. For the present case of analytic input data, we derive derivative growth estimates for the terms of the asymptotic expansion that are explicit in the perturbation parameters and the expansion order.

math.NA↗