arXiv subjects

Cornelia Schneider

Publications and source records attributed to Cornelia Schneider.

At least 19 recordsLinked to original sources

arXiv2026

Approximation classes for the anisotropic space-time finite element method. An almost characterization

We study the approximation of $L_p$-functions, $p\in (0,\infty]$, on cylindrical space-time domains $Ω_T:=[0,T]\times Ω$, $0<T<\infty$, $Ω\subset \R^d$ Lipschitz, $d\in \mathbb{N}$, with respect to continuous anisotropic space-time finite elements on prismatic meshes. In particular, we propose a suitable refinement technique which creates (locally refined) prismatic meshes with sufficient smoothness and the desired anisotropy, and prove complexity estimates. Furthermore, we define a (quasi-)interpolation operator on this type of meshes and use it to characterize the corresponding approximation classes by showing direct and inverse estimates in terms of anisotropic Besov norms.

math.NA↗

arXiv2025

Anisotropic approximation on space-time domains

We investigate anisotropic (piecewise) polynomial approximation of functions in Lebesgue spaces as well as anisotropic Besov spaces. For this purpose we study temporal and spacial moduli of smoothness and their properties. In particular, we prove Jackson- and Whitney-type inequalities on Lipschitz cylinders, i.e., space-time domains $I\times D$ with a finite interval $I$ and a bounded Lipschitz domain $D\subset \R^d$, $d\in \N$. As an application, we prove a direct estimate result for adaptive space-time finite element approximation in the discontinuous setting.

math.NA↗

arXiv2025

Bases of Lebesgue spaces formed by neural networks

The seminal work of Daubechies, DeVore, Foucart, Hanin, and Petrova introduced in 2022 a sequence of univariate piece-wise linear functions, which resemble the classical Fourier basis and which, at the same time, can be easily reproduced by artificial neural networks with ReLU activation function. We give an alternative way how to calculate the inner products of functions from this system and discuss the spectral properties of the Gram matrix generated by this system. The univariate system was later generalized to the multivariate setting by two of the authors of this work. Instead of the usual tensor product construction, this generalization relied on the inner products inside of the argument of the univariate sequence. It turned out that such a system forms a Riesz basis of $L_2(0,1)^n$ for every $n\ge 1$ with Riesz constants independent of $n$. In this work, we investigate the properties of these new sequences of functions in $L_q(0,1)^n$ for $q\not =2.$ First, we show that the univariate system is a Schauder basis in $L_q(0,1)$ for every $1<q<\infty$. By a general argument, it follows that the tensor products of this system also form a Schauder basis in $L_q(0,1)^n$ for every $n\ge 2$ and $1<q<\infty.$ The same fact can also be shown by measuring the distance of the tensor product system to the classical multivariate Fourier basis, but - surprisingly - this argument only works for $n\le 3$. If, on the other hand, we replace the outer tensor products by inner products directly in the argument of the univariate system, the same approach is applicable for an arbitrary dimension $n\in{\mathbb N}.$

math.FA↗

arXiv2025

Nonlocal techniques for the analysis of deep ReLU neural network approximations

Recently, Daubechies, DeVore, Foucart, Hanin, and Petrova introduced a system of piece-wise linear functions, which can be easily reproduced by artificial neural networks with the ReLU activation function and which form a Riesz basis of $L_2([0,1])$. This work was generalized by two of the authors to the multivariate setting. We show that this system serves as a Riesz basis also for Sobolev spaces $W^s([0,1]^d)$ and Barron classes ${\mathbb B}^s([0,1]^d)$ with smoothness $0<s<1$. We apply this fact to re-prove some recent results on the approximation of functions from these classes by deep neural networks. Our proof method avoids using local approximations and allows us to track also the implicit constants as well as to show that we can avoid the curse of dimension. Moreover, we also study how well one can approximate Sobolev and Barron functions by ANNs if only function values are known.

cs.LG↗

arXiv2025

Improved universal approximation with neural networks studied via affine-invariant subspaces of $L_2(\mathbb{R}^n)$

We show that there are no non-trivial closed subspaces of $L_2(\mathbb{R}^n)$ that are invariant under invertible affine transformations. We apply this result to neural networks showing that any nonzero $L_2(\mathbb{R})$ function is an adequate activation function in a one hidden layer neural network in order to approximate every function in $L_2(\mathbb{R})$ with any desired accuracy. This generalizes the universal approximation properties of neural networks in $L_2(\mathbb{R})$ related to Wiener's Tauberian Theorems. Our results extend to the spaces $L_p(\mathbb{R})$ with $p>1$.

math.FA↗

arXiv2025

Besov regularity of non-linear parabolic PDEs on non-convex polyhedral domains

This paper is concerned with the regularity of solutions to parabolic evolution equations. We consider semilinear problems on non-convex domains. Special attention is paid to the smoothness in the specific scale $B^r_{τ,τ}$, $\frac{1}τ=\frac rd+ \frac 1p$ of Besov spaces. The regularity in these spaces determines the approximation order that can be achieved by adaptive and other nonlinear approximation schemes. We show that for all cases under consideration the Besov regularity is high enough to justify the use of adaptive algorithms. Our proofs are based on Schauder's fixed point theorem.

math.AP↗

arXiv2024

Sobolev spaces with mixed weights and the Poisson equation on angular domains

We introduce and analyse a class of weighted Sobolev spaces with mixed weights on angular domains. The weights are based on both the distance to the boundary and the distance to the one vertex of the domain. Moreover, we show how the regularity of the Poisson equation can be analysed in the framework of these spaces by means of the Mellin transform, provided the integrability parameter equals two. Our main motivation comes from the study of stochastic partial differential equations and associated degenerate deterministic parabolic equations.

math.AP↗

arXiv2024

Refined localization spaces, Kondratiev spaces with fractional smoothness and extension operators

In this paper, we introduce Kondratiev spaces of fractional smoothness based on their close relation to refined localization spaces. Moreover, we investigate relations to other approaches leading to extensions of the scale of Kondratiev spaces with integer order of smoothness, based on complex interpolation, and give further results for complex interpolation of those function spaces. As it turns out to be one of the main tools in studying these spaces on domains of polyhedral type, certain aspects of the analysis of Stein's extension operator are revisited. Finally, as an application, we study Sobolev-type embeddings.

math.NA↗

arXiv2024

Relations between Kondratiev spaces and refined localization Triebel-Lizorkin spaces

We investigate the close relation between certain weighted Sobolev spaces (Kondratiev spaces) and refined localization spaces from introduced by Triebel [39,40]. In particular, using a characterization for refined localization spaces from Scharf [32], we considerably improve an embedding from Hansen [17]. This embedding is of special interest in connection with convergence rates for adaptive approximation schemes.

math.AP↗

arXiv2023

A multivariate Riesz basis of ReLU neural networks

We consider the trigonometric-like system of piecewise linear functions introduced recently by Daubechies, DeVore, Foucart, Hanin, and Petrova. We provide an alternative proof that this system forms a Riesz basis of $L_2([0,1])$ based on the Gershgorin theorem. We also generalize this system to higher dimensions $d>1$ by a construction, which avoids using (tensor) products. As a consequence, the functions from the new Riesz basis of $L_2([0,1]^d)$ can be easily represented by neural networks. Moreover, the Riesz constants of this system are independent of $d$, making it an attractive building block regarding future multivariate analysis of neural networks.

cs.IT↗

arXiv2022

Path regularity of Brownian motion and Brownian sheet

By the work of P. Lévy, the sample paths of the Brownian motion are known to satisfy a certain Hölder regularity condition almost surely. This was later improved by Ciesielski, who studied the regularity of these paths in Besov and Besov-Orlicz spaces. We review these results and propose new function spaces of Besov type, strictly smaller than those of Ciesielski and Lévy, where the sample paths of the Brownian motion lie in almost surely. In the same spirit, we review and extend the work of Kamont, who investigated the same question for the multivariate Brownian sheet and function spaces of dominating mixed smoothness.

math.PR↗

arXiv2021

Anisotropic Besov regularity of parabolic PDEs

This paper is concerned with the regularity of solutions to parabolic evolution equations. Special attention is paid to the smoothness in the specific anisotropic scale $\ B^{r\mathbf{a}}_{τ,τ}, \ \frac{1}τ=\frac{r}{d}+\frac{1}{p}\ $ of Besov spaces where $\mathbf{a}$ measures the anisotropy. The regularity in these spaces determines the approximation order that can be achieved by fully space-time adaptive approximation schemes. In particular, we show that for the heat equation our results significantly improve previous results by Aimar and Gomez [3].

math.AP↗

arXiv2021

Regularity in Sobolev and Besov spaces for parabolic problems on domains of polyhedral type

This paper is concerned with the regularity of solutions to linear and nonlinear evolution equations extending our findings in [22] to domains of polyhedral type. In particular, we study the smoothness in the specific scale $B^r_{τ,τ}$, $\frac{1}τ=\frac rd+\frac 1p$ of Besov spaces. The regularity in these spaces determines the approximation order that can be achieved by adaptive and other nonlinear approximation schemes. We show that for all cases under consideration the Besov regularity is high enough to justify the use of adaptive algorithms.

math.AP↗

arXiv2021

Sobolev meets Besov: Regularity for the Poisson equation with Dirichlet, Neumann and mixed boundary values

We study the regularity of solutions of the Poisson equation with Dirichlet, Neumann and mixed boundary values in polyhedral cones $K\subset \mathbb{R}^3$ in the specific scale $\ B^α_{τ,τ}, \ \frac{1}τ=\fracα{d}+\frac{1}{p}\ $ of Besov spaces. The regularity of the solution in these spaces determines the order of approximation that can be achieved by adaptive and nonlinear numerical schemes. We aim for a thorough discussion of homogeneous and inhomogeneous boundary data in all settings studied and show that the solutions are much smoother in this specific Besov scale compared to the fractional Sobolev scale $H^s$ in all cases, which justifies the use of adaptive schemes.

math.AP↗

arXiv2021

Approximation classes for adaptive time-stepping finite element methods

We study approximation classes for adaptive time-stepping finite element methods for time-dependent Partial Differential Equations (PDE). We measure the approximation error in $L_2([0,T)\timesΩ)$ and consider the approximation with discontinuous finite elements in time and continuous finite elements in space, of any degree. As a byproduct we define Besov spaces for vector-valued functions on an interval and derive some embeddings, as well as Jackson- and Whitney-type estimates.

math.NA↗

arXiv2021

An extension operator for Sobolev spaces with mixed weights

We provide an extension operator for weighted Sobolev spaces on bounded polyhedral cones $K$ involving a mixture of weights, which measure the distance to the vertex and the edges of the cone, respectively. Our results are based on Stein's extension operator for Sobolev spaces.

math.FA↗

arXiv2020

Growth envelopes of some variable and mixed function spaces

We study unboundedness properties of functions belonging Lebesgue and Lorentz spaces with variable and mixed norms using growth envelopes. Our results extend the ones for the corresponding classical spaces in a natural way. In the case of spaces with mixed norms it turns out that the unboundedness in the worst direction, i.e., in the direction where $p_{i}$ is the smallest, is crucial. More precisely, the growth envelope is given by $E_G(L_{\vec{p}}(Ω)) = (t^{-1/\min\{p_{1}, \ldots, p_{d} \}},\min\{p_{1}, \ldots, p_{d} \})$ for mixed Lebesgue and $E_G(L_{\vec{p},q}(Ω)) = (t^{-1/\min\{p_{1}, \ldots, p_{d} \}},q)$ for mixed Lorentz spaces, respectively. For the variable Lebesgue spaces we obtain $E_G(L_{p(\cdot)}(Ω)) = (t^{-1/p_{-}},p_{-})$, where $p_{-}$ is the essential infimum of $p(\cdot)$, subject to some further assumptions. Similarly, for the variable Lorentz space it holds $E_G(L_{p(\cdot),q}(Ω)) = (t^{-1/p_{-}},q)$. The growth envelope is used for Hardy-type inequalities and limiting embeddings. In particular, as a by-product we determine the smallest classical Lebesgue (Lorentz) space which contains a fixed mixed or variable Lebesgue (Lorentz) space, respectively.

math.FA↗

arXiv2019

Properties of Kondratiev spaces

In this paper, we investigate Kondratiev spaces on domains of polyhedral type. In particular, we will be concerned with necessary and sufficient conditions for continuous and compact embeddings, and in addition we shall deal with pointwise multiplication in these spaces.

math.FA↗