46
92

Deep ReLU network approximation of functions on a manifold

Abstract

Whereas recovery of the manifold from data is a well-studied topic, approximation rates for functions defined on manifolds are less known. In this work, we study a regression problem with inputs on a dd^*-dimensional manifold that is embedded into a space with potentially much larger ambient dimension. It is shown that sparsely connected deep ReLU networks can approximate a H\"older function with smoothness index β\beta up to error ϵ\epsilon using of the order of ϵd/βlog(1/ϵ)\epsilon^{-d^*/\beta}\log(1/\epsilon) many non-zero network parameters. As an application, we derive statistical convergence rates for the estimator minimizing the empirical risk over all possible choices of bounded network parameters.

View on arXiv
Comments on this paper