Universal, sample-optimal algorithms for recovery of anisotropic functions from i.i.d. samples

TL;DR

This paper develops universal, nonadaptive algorithms based on compressed sensing for high-dimensional anisotropic function recovery, achieving near-optimal rates and demonstrating the limitations of linear methods.

Contribution

It introduces a universal, nonlinear algorithm for anisotropic function recovery that is nearly optimal and shows linear algorithms are inherently suboptimal.

Findings

The proposed algorithm recovers anisotropic functions with near-optimal rates.

Linear algorithms are shown to be suboptimal due to the curse of dimensionality.

Lower bounds establish the near-optimality of the nonlinear approach.

Abstract

A key problem in approximation theory is the recovery of high-dimensional functions from samples. In many cases, the functions of interest exhibit anisotropic smoothness, and, in many practical settings, the nature of this anisotropy may be unknown a priori. Therefore, an important question involves the development of universal algorithms, namely, algorithms that simultaneously achieve optimal or near-optimal rates of convergence across a range of different anisotropic smoothness classes. In this work, we consider universal approximation of periodic functions that belong to anisotropic Sobolev spaces and anisotropic dominating mixed smoothness Sobolev spaces. Our first result is the construction of a universal algorithm. This recasts function recovery as a sparse recovery problem for Fourier coefficients and then exploits compressed sensing to yield the desired approximation rates. Note that this algorithm is nonadaptive, as it does not seek to learn the anisotropic smoothness of the target function. We then demonstrate optimality of this algorithm up to a dimension-independent polylogarithmic factor. We do this by presenting a lower bound for the adaptive $m$-width for the unit balls of such function classes. Finally, we demonstrate the necessity of nonlinear algorithms. We show that universal linear algorithms can achieve rates that are at best suboptimal by a dimension-dependent polylogarithmic factor. In other words, they suffer from a curse of dimensionality in the rate -- a phenomenon which justifies the necessity of nonlinear algorithms for universal recovery.