Saved in:
Bibliographic Details
Main Author: Sellke, Mark
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2306.01992
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910318142160896
author Sellke, Mark
author_facet Sellke, Mark
contents We study the sample complexity of learning ReLU neural networks from the point of view of generalization. Given norm constraints on the weight matrices, a common approach is to estimate the Rademacher complexity of the associated function class. Previously Golowich-Rakhlin-Shamir (2020) obtained a bound independent of the network size (scaling with a product of Frobenius norms) except for a factor of the square-root depth. We give a refinement which often has no explicit depth-dependence at all.
format Preprint
id arxiv_https___arxiv_org_abs_2306_01992
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle On Size-Independent Sample Complexity of ReLU Networks
Sellke, Mark
Machine Learning
We study the sample complexity of learning ReLU neural networks from the point of view of generalization. Given norm constraints on the weight matrices, a common approach is to estimate the Rademacher complexity of the associated function class. Previously Golowich-Rakhlin-Shamir (2020) obtained a bound independent of the network size (scaling with a product of Frobenius norms) except for a factor of the square-root depth. We give a refinement which often has no explicit depth-dependence at all.
title On Size-Independent Sample Complexity of ReLU Networks
topic Machine Learning
url https://arxiv.org/abs/2306.01992