A Class of Function-Preserving Neural Network Operators with Applications to Modulated Signal Denoising

Sachin Saini

Home/Archive/arXiv

arXiv · 2610.05759

A Class of Function-Preserving Neural Network Operators with Applications to Modulated Signal Denoising

Sachin Saini

View original source ↗Read source document ↗Export metadata ↓

Abstract

In this paper, we introduce a class of function-preserving neural network operators designed to preserve a prescribed strictly positive function. We examine the approximation behaviour of the proposed operators on the space of continuous functions and show that they are positive, linear, well-defined, and bounded. Specifically, we demonstrate uniform convergence on compact intervals and get quantitative estimates of the approximation error in terms of the modulus of continuity. To support the theoretical analysis, we present several numerical experiments focusing on the preservation of exponential and logarithmic functions. These examples show how well the proposed operators perform and compare them with existing neural network operators. Also, an application to denoising modulated signals is provided, demonstrating the practical relevance of the proposed approach.

Explore related subjects

math.FA cs.NA math.NA

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Sachin Saini. 2026-10-05. A Class of Function-Preserving Neural Network Operators with Applications to Modulated Signal Denoising. https://arxiv.org/abs/2610.05759

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Finitely $C^\infty$-generated associative and Hopf algebras

We introduce finitely $C^\infty$-generated algebras, which can be treated as `algebras of functions' on non-commutative $C^\infty$-differentiable spaces. Our approach uses the category of projective limits of real Banach algebras of polynomial growth. We prove the existence of some universal constructions in this and some similar categories. By analogy with holomorphically finitely generated algebras of Pirkovskii, a finitely $C^\infty$-generated algebra is defined as a quotient of a finite-rank algebra of `free $C^\infty$-functions'. The latter notion was introduced by the author in a previous article, where a structure theorem for algebras of `free $C^\infty$-functions' was announced and proved in dimension at most $2$. Here this theorem is proved in full generality. The central result asserts that the projective tensor product of a finite tuple of finitely $C^\infty$-generated algebras is finitely $C^\infty$-generated. In particular, this makes it natural to consider finitely $C^\infty$-generated topological Hopf algebras. Furthermore, a construction called `envelope' provides a functor from the category of affine real Hopf algebras to the category of finitely $C^\infty$-generated Hopf algebras.

math.FA↗

arXiv2026

Composition-differentiation operators on Hardy-Hilbert space of Dirichlet series

In this paper, we establish a compactness criterion for the composition-differentiation operator $D_Φ$ on the Hardy space $\mathcal{H}^2$ of Dirichlet series in terms of a boundary decay condition for its mean counting function. Via a comparison-type principle, we show that this boundary decay is equivalent to a corresponding decay condition for the Green's function, which we further analyze through harmonic measure. We provide explicit mapping properties of the symbol $Φ$ that generate a bounded composition-differentiation operator $D_Φ$ and obtain precise norm estimates for $D_Φ$ when $Φ$ is an affine symbol with a single-prime in the class $\mathcal{G}_0$. Furthermore, we establish explicit upper and lower bounds for the approximation numbers of $D_Φ$ on $\mathcal{H}^2$ motivated by the work of Queffélec and Seip [J. Funct. Anal., 2015]. Finally, we investigate spectral and operator-theoretic properties of $D_Φ$ for symbols in $\mathcal{G}_0$.

math.FA↗

arXiv2026

Sobolev spaces in infinite dimensions

The classical theory of Sobolev spaces in finite dimensions is well established. Because infinite-dimensional spaces possess inherent analytical and topological complexities, developing a theory of Sobolev spaces for functions of infinitely many variables---rather than merely extending the classical finite-dimensional theory---is far from routine and has led to long-standing open problems. In this paper, we establish such a theory and systematically determine the extent to which these classical results can be carried over to this setting. By carefully adapting the tools introduced in our recent works, we establish a series of infinite-dimensional counterparts of the classical theorems and, in the process, reveal new phenomena with no finite-dimensional analogue. The methods and concepts developed here, together with the results obtained, furnish a robust framework for further study of Sobolev spaces and related problems in infinite-dimensional analysis.

math.FA↗