Towards Feature Space Adversarial Attack by Style Perturbation

Authors

  • Qiuling Xu Purdue University
  • Guanhong Tao Purdue University
  • Siyuan Cheng Purdue University
  • Xiangyu Zhang Purdue University

Keywords:

Adversarial Learning & Robustness

Abstract

We propose a new adversarial attack to Deep Neural Networks for image classification. Different from most existing attacks that directly perturb input pixels, our attack focuses on perturbing abstract features, more specifically, features that denote styles, including interpretable styles such as vivid colors and sharp outlines, and uninterpretable ones. It induces model misclassification by injecting imperceptible style changes through an optimization procedure. We show that our attack can generate adversarial samples that are more natural-looking than the state-of-the-art unbounded attacks. The experiment also supports that existing pixel-space adversarial attack detection and defense techniques can hardly ensure robustness in the style-related feature space.

Downloads

Published

2021-05-18

How to Cite

Xu, Q., Tao, G., Cheng, S., & Zhang, X. (2021). Towards Feature Space Adversarial Attack by Style Perturbation. Proceedings of the AAAI Conference on Artificial Intelligence, 35(12), 10523-10531. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/17259

Issue

Section

AAAI Technical Track on Machine Learning V