Disentangling Adversarial Robustness and Generalization

Obtaining deep networks that are robust against adversarial examples and generalize well is an open problem. A recent hypothesis even states that both robust and accurate models are impossible, i.e., adversarial robustness and generalization are conflicting goals. In an effort to clarify the relatio...

Full description

Saved in:
Bibliographic Details
Published in2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) pp. 6969 - 6980
Main Authors Stutz, David, Hein, Matthias, Schiele, Bernt
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.06.2019
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Obtaining deep networks that are robust against adversarial examples and generalize well is an open problem. A recent hypothesis even states that both robust and accurate models are impossible, i.e., adversarial robustness and generalization are conflicting goals. In an effort to clarify the relationship between robustness and generalization, we assume an underlying, low-dimensional data manifold and show that: 1. regular adversarial examples leave the manifold; 2. adversarial examples constrained to the manifold, i.e., on-manifold adversarial examples, exist; 3. on-manifold adversarial examples are generalization errors, and on-manifold adversarial training boosts generalization; 4. regular robustness and generalization are not necessarily contradicting goals. These assumptions imply that both robust and accurate models are possible. However, different models (architectures, training strategies etc.) can exhibit different robustness and generalization characteristics. To confirm our claims, we present extensive experiments on synthetic data (with known manifold) as well as on EMNIST, Fashion-MNIST and CelebA.
ISSN:2575-7075
DOI:10.1109/CVPR.2019.00714