Optimal learning of high-dimensional classification problems using deep neural networks
Functional Analysis
2021-12-28 v2 Machine Learning
Machine Learning
Abstract
We study the problem of learning classification functions from noiseless training samples, under the assumption that the decision boundary is of a certain regularity. We establish universal lower bounds for this estimation problem, for general classes of continuous decision boundaries. For the class of locally Barron-regular decision boundaries, we find that the optimal estimation rates are essentially independent of the underlying dimension and can be realized by empirical risk minimization methods over a suitable class of deep neural networks. These results are based on novel estimates of the and entropies of the class of Barron-regular functions.
Cite
@article{arxiv.2112.12555,
title = {Optimal learning of high-dimensional classification problems using deep neural networks},
author = {Philipp Petersen and Felix Voigtlaender},
journal= {arXiv preprint arXiv:2112.12555},
year = {2021}
}