Faculty, Staff and Student Publications

Language

English

Publication Date

1-1-2024

Journal

Proceedings of Machine Learning Research

PMID

39206101

PMCID

PMC11350389

PubMedCentral® Posted Date

8-28-2024

PubMedCentral® Full Text Version

Author MSS

Abstract

Training Deep Neural Networks (DNNs) with adversarial examples often results in poor generalization to test-time adversarial data. This paper investigates this issue, known as adversarially robust generalization, through the lens of Rademacher complexity. Building upon the studies by Khim and Loh (2018); Yin et al. (2019), numerous works have been dedicated to this problem, yet achieving a satisfactory bound remains an elusive goal. Existing works on DNNs either apply to a surrogate loss instead of the robust loss or yield bounds that are notably looser compared to their standard counterparts. In the latter case, the bounds have a higher dependency on the width𝑚 of the DNNs or the dimension𝑑 of the data, with an extra factor of at least𝒪⁡(√𝑚) or𝒪⁡(√𝑑).

This paper presents upper bounds for adversarial Rademacher complexity of DNNs that match the best-known upper bounds in standard settings, as established in the work of Bartlett et al. (2017), with the dependency on width and dimension being 𝒪⁡(ln ⁡(𝑑⁢𝑚)). The central challenge addressed is calculating the covering number of adversarial function classes. We aim to construct a new cover that possesses two properties: 1) compatibility with adversarial examples, and 2) precision comparable to covers used in standard settings. To this end, we introduce a new variant of covering number called the uniform covering number, specifically designed and proven to reconcile these two properties. Consequently, our method effectively bridges the gap between Rademacher complexity in robust and standard generalization.

Keywords

Adversarially Robust Generalization, Rademacher Complexity, Covering Number

Published Open-Access

yes

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.