English

Deep ReLU networks overcome the curse of dimensionality for bandlimited functions

Numerical Analysis 2020-11-10 v4 Numerical Analysis

Abstract

We prove a theorem concerning the approximation of bandlimited multivariate functions by deep ReLU networks for which the curse of the dimensionality is overcome. Our theorem is based on a result by Maurey and on the ability of deep ReLU networks to approximate Chebyshev polynomials and analytic functions efficiently.

Keywords

Cite

@article{arxiv.1903.00735,
  title  = {Deep ReLU networks overcome the curse of dimensionality for bandlimited functions},
  author = {Hadrien Montanelli and Haizhao Yang and Qiang Du},
  journal= {arXiv preprint arXiv:1903.00735},
  year   = {2020}
}