51
34

Besov Function Approximation and Binary Classification on Low-Dimensional Manifolds Using Convolutional Residual Networks

Abstract

Most of existing statistical theories on deep neural networks have sample complexities cursed by the data dimension and therefore cannot well explain the empirical success of deep learning on high-dimensional data. To bridge this gap, we propose to exploit low-dimensional geometric structures of the real world data sets. We establish theoretical guarantees of convolutional residual networks (ConvResNet) in terms of function approximation and statistical estimation for binary classification. Specifically, given the data lying on a dd-dimensional manifold isometrically embedded in RD\mathbb{R}^D, we prove that if the network architecture is properly chosen, ConvResNets can (1) approximate Besov functions on manifolds with arbitrary accuracy, and (2) learn a classifier by minimizing the empirical logistic risk, which gives an excess risk in the order of ns2s+2(sd)n^{-\frac{s}{2s+2(s\vee d)}}, where ss is a smoothness parameter. This implies that the sample complexity depends on the intrinsic dimension dd, instead of the data dimension DD. Our results demonstrate that ConvResNets are adaptive to low-dimensional structures of data sets.

View on arXiv
Comments on this paper