Super-resolution meets machine learning: approximation of measures

DOI10.1007/S00041-019-09693-XMaRDI QIDQ2338563zbMATH OpenOpenAlexWikidataFDO

Authors H. N. Mhaskar

Publication date 21 November 2019

Published in The Journal of Fourier Analysis and Applications (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1907.04895

zbMATH Keywords

data defined spacesdeconvolutionmachine learningsuper-resolutionwidths

Mathematics Subject Classification ID

Rate of convergence, degree of approximation (41A25) Numerical solution to inverse problems in abstract spaces (65J22) Learning and adaptive systems in artificial intelligence (68T05) Signal theory (characterization, reconstruction, filtering, etc.) (94A12)

Abstract: The problem of super-resolution in general terms is to recuperate a finitely supported measure $mu$ given finitely many of its coefficients $hatmu(k)$ with respect to some orthonormal system. The interesting case concerns situations, where the number of coefficients required is substantially smaller than a power of the reciprocal of the minimal separation among the points in the support of $mu$. In this paper, we consider the more severe problem of recuperating $mu$ approximately without any assumption on $mu$ beyond having a finite total variation. In particular, $mu$ may be supported on a continuum, so that the minimal separation among the points in the support of $mu$ is $0$. A variant of this problem is also of interest in machine learning as well as the inverse problem of de-convolution. We define an appropriate notion of a distance between the target measure and its recuperated version, give an explicit expression for the recuperation operator, and estimate the distance between $mu$ and its approximation. We show that these estimates are the best possible in many different ways. We also explain why for a finitely supported measure the approximation quality of its recuperation is bounded from below if the amount of information is smaller than what is demanded in the super-resolution problem.

Recommendations

• Approximate super-resolution of positive measures in all dimensions
• Towards a Mathematical Theory of Super‐resolution
• Sets of Superresolution and the Maximum Entropy Method on the Mean
• Multivariate super-resolution without separation
• Super-resolution by means of Beurling minimal extrapolation

Cites work

• A quadrature formula for diffusion polynomials corresponding to a generalized heat kernel
• A unified framework for harmonic analysis of functions on directed graphs and changing data
• Accurate Reconstructions of Functions of Finite Regularity from Truncated Fourier Series Expansions
• Adaptive filters for piecewise smooth spectral data*
• Bounds on rates of variable-basis and neural-network approximation
• Comparison of worst case errors in linear and neural network approximation
• Deep vs. shallow networks: an approximation theory perspective
• Detection of edges in spectral data
• Diffusion polynomial frames on metric measure spaces
• Eignets for function approximation on manifolds
• Exact recovery of Dirac ensembles from the projection onto spaces of spherical harmonics
• scientific article; zbMATH DE number 5797591 (Why is no real title available?)
• scientific article; zbMATH DE number 3179793 (Why is no real title available?)
• scientific article; zbMATH DE number 3477793 (Why is no real title available?)
• scientific article; zbMATH DE number 477682 (Why is no real title available?)
• scientific article; zbMATH DE number 1881986 (Why is no real title available?)
• Marcinkiewicz-Zygmund measures on manifolds
• On local smoothness classes of periodic functions
• On the accuracy of solving confluent Prony systems
• On the problem of parameter estimation in exponential sums
• On the tractability of multivariate integration and approximation by neural networks
• Optimal nonlinear approximation
• Parameter estimation for exponential sums by approximate prony method
• Stability and super-resolution of generalized spike recovery
• Super-resolution from noisy data
• Super-Resolution on the Sphere Using Convex Optimization
• Superresolution via Sparsity Constraints
• Towards a Mathematical Theory of Super‐resolution
• Tractability of multivariate problems. Volume I: Linear information

Cited in

(5)

• Approximate super-resolution of positive measures in all dimensions
• Applied harmonic analysis and data science. Abstracts from the workshop held November 28 -- December 4, 2021 (hybrid meeting)
• Resolution Enhancement in $\Sigma\Delta$ Learners for Superresolution Source Separation
• Approximation and interpolation of singular measures by trigonometric polynomials
• A signal separation view of classification

This page was built for publication: Super-resolution meets machine learning: approximation of measures

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2338563)