Source author record

Yurii Kolomoitsev

Explore this author’s publications, collaborators and research topics. This catalog record has not been claimed by its author.

ResearcherUnclaimed source record

math.CA math.NA Numerical Analysis math.FA

Catalog footprint

Research overview

13works

4topics

4close collaborators

Actions

Is this your work?

Open graph Browse papers

preprint2022arXiv

Sharp $L_p$ -error estimates for sampling operators

We study approximation properties of linear sampling operators in the spaces $L_p$ for $1\le p<\infty$ . By means of the Steklov averages, we introduce a new measure of smoothness that simultaneously contains information on the smoothness of a function in $L_p$ and discrete information on the behaviour of a function at sampling points. The new measure of smoothness enables us to improve and extend several classical results of approximation theory to the case of linear sampling operators. In particular, we obtain matching direct and inverse approximation inequalities for sampling operators in $L_p$ , find the exact order of decay of the corresponding $L_p$ -errors for particular classes of functions, and introduce a special $K$ -functional and its realization suitable for studying smoothness properties of sampling operators.

preprint2022arXiv

Spline Representation and Redundancies of One-Dimensional ReLU Neural Network Models

We analyze the structure of a one-dimensional deep ReLU neural network (ReLU DNN) in comparison to the model of continuous piecewise linear (CPL) spline functions with arbitrary knots. In particular, we give a recursive algorithm to transfer the parameter set determining the ReLU DNN into the parameter set of a CPL spline function. Using this representation, we show that after removing the well-known parameter redundancies of the ReLU DNN, which are caused by the positive scaling property, all remaining parameters are independent. Moreover, we show that the ReLU DNN with one, two or three hidden layers can represent CPL spline functions with $K$ arbitrarily prescribed knots (breakpoints), where $K$ is the number of real parameters determining the normalized ReLU DNN (up to the output layer parameters). Our findings are useful to fix a priori conditions on the ReLU DNN to achieve an output with prescribed breakpoints and function values.

preprint2020arXiv

Approximation by multivariate quasi-projection operators and Fourier multipliers

Multivariate quasi-projection operators $Q_j(f,φ, \widetildeφ)$ , associated with a function $φ$ and a distribution/function $\widetildeφ$ , are considered. The function $φ$ is supposed to satisfy the Strang-Fix conditions and a compatibility condition with $\widetildeφ$ . Using technique based on the Fourier multipliers, we studied approximation properties of such operators for functions $f$ from anisotropic Besov spaces and $L_p$ spaces with $1\le p\le \infty$ . In particular, upper and lower estimates of the $L_p$ -error of approximation in terms of moduli of smoothness and best approximations are obtained.

preprint2020arXiv

Approximation by sampling-type operators in $L_p$ -spaces

Approximation properties of the sampling-type quasi-projection operators $Q_j(f,φ, \widetildeφ)$ for functions $f$ from anisotropic Besov spaces are studied. Error estimates in $L_p$ -norm are obtained for a large class of tempered distributions $\widetildeφ$ and a large class of functions $φ$ under the assumptions that $φ$ has enough decay, satisfies the Strang-Fix conditions and a compatibility condition with $\widetildeφ$ . The estimates are given in terms of moduli of smoothness and best approximations.

preprint2020arXiv

Asymptotics of the Lebesgue constants for a $d$ -dimensional simplex

In this paper an asymptotic formula is given for the Lebesgue constants generated by the anisotropically dilated $d$ -dimensional simplex. Contrary to many preceding results established only in dimension two, the obtained ones are proved in any dimension. Also, the "rational" and "irrational" parts are both united and separated in one formula.

preprint2020arXiv

Asymptotics of the Lebesgue constants for bivariate approximation processes

In this paper asymptotic formulas are given for the Lebesgue constants generated by three special approximation processes related to the $\ell_1$ -partial sums of Fourier series. In particular, we consider the Lagrange interpolation polynomials based on the Lissajous-Chebyshev node points, the partial sums of the Fourier series generated by the anisotropically dilated rhombus, and the corresponding discrete partial sums.

preprint2020arXiv

Properties of moduli of smoothness in $L_p(\mathbb{R}^d)$

In this paper, we discuss various basic properties of moduli of smoothness of functions from $L_p(\mathbb{R}^d)$ , $0<p\le \infty$ . In particular, complete versions of Jackson-, Marchaud-, and Ulyanov-type inequalities are given for the whole range of $p$ . Moreover, equivalences between moduli of smoothness and the corresponding $K$ -functionals and the realization concept are proved.

preprint2020arXiv

Quasi-projection operators in the weighted $L_p$ spaces

Approximation properties of multivariate quasi-projection operators are studied in the paper. Wide classes of such operators are considered, including the sampling and the Kantorovich-Kotelnikov type operators generated by different band-limited functions.The rate of convergence in the weighted $L_p$ -spaces for these operators is investigated. The results allow to estimate the error for reconstruction of signals (approximated functions) whose decay is not enough to be in $L_p$ .

preprint2020arXiv

Uniform approximation by multivariate quasi-projection operators

Approximation properties of quasi-projection operators $Q_j(f,φ, \widetildeφ)$ are studied. Such an operator is associated with a function $φ$ satisfying the Strang-Fix conditions and a tempered distribution $\widetildeφ$ such that compatibility conditions with $φ$ hold. Error estimates in the uniform norm are obtained for a wide class of quasi-projection operators defined on the space of uniformly continuous functions and on the anisotropic Besov spaces. Under additional assumptions on $φ$ and $\widetildeφ$ , two-sided estimates in terms of realizations of the $K$ -functional are also obtained.

preprint2016arXiv

Best approximations and moduli of smoothness of functions and their derivatives in $L_p$ , $0<p<1$

Several new inequalities for moduli of smoothness and errors of the best approximation of a function and its derivatives in the spaces $L_p$ , $0<p<1$ , are obtained. For example, it is shown that for any $0<p<1$ and $k,\,r\in \mathbb{N}$ one has $ ω_{r+k}(f,\d)_p\leq C({p,k,r})\d^{r+\frac{1}{p}-1}$\int_0^\d\frac{ω_{k}(f^{(r)},t)_p^p}{t^{2-p}}{\rm d}t$^\frac{1}{p},$ where the function $f$ is such that $f^{(r-1)}$ is absolutely continuous. Similar inequalities are obtained for the Ditzian-Totik moduli of smoothness and the error of the best approximation of functions by trigonometric and algebraic polynomials and splines. As an application, positive results about simultaneous approximation of a function and its derivatives by the mentioned approximation methods in the spaces $L_p$ , $0<p<1$ , are derived.

preprint2016arXiv

On approximation of functions by algebraic polynomials in Hölder spaces

We study approximation of functions by algebraic polynomials in the Hölder spaces corresponding to the generalized Jacobi translation and the Ditzian-Totik moduli of smoothness. By using modifications of the classical moduli of smoothness, we give improvements of the direct and inverse theorems of approximation and prove the criteria of the precise order of decrease of the best approximation in these spaces. Moreover, we obtain strong converse inequalities for some methods of approximation of functions. As an example, we consider approximation by the Durrmeyer-Bernstein polynomial operators.

preprint2015arXiv

Sharp estimates of approximation of periodic functions in Hölder spaces

The main purpose of the paper is to study sharp estimates of approximation of periodic functions in the Hölder spaces $H_p^{r,α}$ for all $0<p\le\infty$ and $0<α\le r$ . By using modifications of the classical moduli of smoothness, we give improvements of the direct and inverse theorems of approximation and prove the criteria for the precise order of decrease of the best approximation in these spaces. Moreover, we obtained strong converse inequalities for general methods of approximation of periodic functions in $H_p^{r,α}$ .

preprint2011arXiv

On approximation methods generated by generalized Bochner-Riesz kernels

Some sharp results related to the convergence of means and families of operators generated by the generalized Bochner-Riesz kernels are obtained. The exact order of approximation of functions by these methods via $K$ -functional (or its realization in the case of the space $L_p$ , $0<p<1$ ) is derived.

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Yurii Kolomoitsev

Research overview

Is this your work?

13 publications

Sharp $L_p$ -error estimates for sampling operators

Spline Representation and Redundancies of One-Dimensional ReLU Neural Network Models

Approximation by multivariate quasi-projection operators and Fourier multipliers

Approximation by sampling-type operators in $L_p$ -spaces

Asymptotics of the Lebesgue constants for a $d$ -dimensional simplex

Asymptotics of the Lebesgue constants for bivariate approximation processes

Properties of moduli of smoothness in $L_p(\mathbb{R}^d)$

Quasi-projection operators in the weighted $L_p$ spaces

Uniform approximation by multivariate quasi-projection operators

Best approximations and moduli of smoothness of functions and their derivatives in $L_p$ , $0<p<1$

On approximation of functions by algebraic polynomials in Hölder spaces

Sharp estimates of approximation of periodic functions in Hölder spaces

On approximation methods generated by generalized Bochner-Riesz kernels

See the researcher in context

Building this map preview

Yurii Kolomoitsev

Research overview

Is this your work?

13 publications

Sharp LpL_pLp​-error estimates for sampling operators

Spline Representation and Redundancies of One-Dimensional ReLU Neural Network Models

Approximation by multivariate quasi-projection operators and Fourier multipliers

Approximation by sampling-type operators in LpL_pLp​-spaces

Asymptotics of the Lebesgue constants for a ddd-dimensional simplex

Asymptotics of the Lebesgue constants for bivariate approximation processes

Properties of moduli of smoothness in Lp(Rd)L_p(\mathbb{R}^d)Lp​(Rd)

Quasi-projection operators in the weighted LpL_pLp​ spaces

Uniform approximation by multivariate quasi-projection operators

Best approximations and moduli of smoothness of functions and their derivatives in LpL_pLp​, 0<p<10<p<10<p<1

On approximation of functions by algebraic polynomials in Hölder spaces

Sharp estimates of approximation of periodic functions in Hölder spaces

On approximation methods generated by generalized Bochner-Riesz kernels

See the researcher in context

Building this map preview

Sharp $L_p$ -error estimates for sampling operators

Approximation by sampling-type operators in $L_p$ -spaces

Asymptotics of the Lebesgue constants for a $d$ -dimensional simplex

Properties of moduli of smoothness in $L_p(\mathbb{R}^d)$

Quasi-projection operators in the weighted $L_p$ spaces

Best approximations and moduli of smoothness of functions and their derivatives in $L_p$ , $0<p<1$