Understanding adversarial robustness on the hypersphere

Zitian Zhao

doi:10.1117/12.2681304

8 June 2023 Understanding adversarial robustness on the hypersphere

Zitian Zhao

Proceedings Volume 12707, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023); 127072F (2023) https://doi.org/10.1117/12.2681304
Event: International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023), 2023, Changsha, China

Abstract

Adversarial examples have raised public concern about the robustness of deep neural networks (DNNs). One universal approach to enhance the robustness is adversarial training which essentially augments the training data. However, adversarial training succeeded only to a very limited extent. This limited progress is partly due to the lack of interpretation and understanding of the robustness of DNNs. In this context, we try to explain the adversarial robustness by embedding the sample points onto a hypersphere, which naturally provides an interpretable metric for the distance between sample points. Different from the empirical intuition, it is observed that adversarial trained models show complex patterns when facing different datasets and training configurations. Observations and explanations about robustness and model behavior are made from the aspect of the distances between samples points. Lastly, we discuss the degradation of standard accuracy in the adversarial trained models and provide possible solutions.

Citation Download Citation

Zitian Zhao "Understanding adversarial robustness on the hypersphere", Proc. SPIE 12707, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023), 127072F (8 June 2023); https://doi.org/10.1117/12.2681304

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
8 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Adversarial training

Data modeling

Feature extraction

Statistical modeling

Convolution

Distance measurement

Neural networks

Show All Keywords

Keywords/Phrases

Search In:

Publication Years