A general learning algorithm for functions between metric spaces

Abstract

In this paper we show how to approximate ("learn") a function f, where X and Y are metric spaces.