ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2202.06392
19
8

Local approximation of operators

13 February 2022
H. Mhaskar
ArXivPDFHTML
Abstract

Many applications, such as system identification, classification of time series, direct and inverse problems in partial differential equations, and uncertainty quantification lead to the question of approximation of a non-linear operator between metric spaces X\mathfrak{X}X and Y\mathfrak{Y}Y. We study the problem of determining the degree of approximation of such operators on a compact subset KX⊂XK_\mathfrak{X}\subset \mathfrak{X}KX​⊂X using a finite amount of information. If F:KX→KY\mathcal{F}: K_\mathfrak{X}\to K_\mathfrak{Y}F:KX​→KY​, a well established strategy to approximate F(F)\mathcal{F}(F)F(F) for some F∈KXF\in K_\mathfrak{X}F∈KX​ is to encode FFF (respectively, F(F)\mathcal{F}(F)F(F)) in terms of a finite number ddd (repectively mmm) of real numbers. Together with appropriate reconstruction algorithms (decoders), the problem reduces to the approximation of mmm functions on a compact subset of a high dimensional Euclidean space Rd\mathbb{R}^dRd, equivalently, the unit sphere Sd\mathbb{S}^dSd embedded in Rd+1\mathbb{R}^{d+1}Rd+1. The problem is challenging because ddd, mmm, as well as the complexity of the approximation on Sd\mathbb{S}^dSd are all large, and it is necessary to estimate the accuracy keeping track of the inter-dependence of all the approximations involved. In this paper, we establish constructive methods to do this efficiently; i.e., with the constants involved in the estimates on the approximation on Sd\mathbb{S}^dSd being O(d1/6)\mathcal{O}(d^{1/6})O(d1/6). We study different smoothness classes for the operators, and also propose a method for approximation of F(F)\mathcal{F}(F)F(F) using only information in a small neighborhood of FFF, resulting in an effective reduction in the number of parameters involved.

View on arXiv
Comments on this paper