Convergence rates of vector-valued local polynomial regression

Publication date 13 July 2021

Abstract: Non-parametric estimation of functions as well as their derivatives by means of local-polynomial regression is a subject that was studied in the literature since the late 1970's. Given a set of noisy samples of a

m a t h c a l C^{k}

smooth function, we perform a local polynomial fit, and by taking its

m

-th derivative we obtain an estimate for the

m

-th function derivative. The known optimal rates of convergence for this problem for a

k

-times smooth function

f : m a t h b b R^{d} o m a t h b b R

are

n^{- f r a c k - m 2 k + d}

. However in modern applications it is often the case that we have to estimate a function operating to

m a t h b b R^{D}

, for

D g g d

extremely large. In this work, we prove that these same rates of convergence are also achievable by local-polynomial regression in case of a high dimensional target, given some assumptions on the noise distribution. This result is an extension to Stone's seminal work from 1980 to the regime of high-dimensional target domain. In addition, we unveil a connection between the failure probability

v a r e p s i l o n

and the number of samples required to achieve the optimal rates.

This page was built for publication: Convergence rates of vector-valued local polynomial regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6372638)