Affiliation:
1. Carnegie Mellon University, Pittsburgh, PA, USA
2. University of North Carolina at Chapel Hill, North Carolina, USA
Abstract
Images perturbed subtly to be misclassified by neural networks, called
adversarial examples
, have emerged as a technically deep challenge and an important concern for several application domains. Most research on adversarial examples takes as its only constraint that the perturbed images are similar to the originals. However, real-world application of these ideas often requires the examples to satisfy additional objectives, which are typically enforced through custom modifications of the perturbation process. In this article, we propose
adversarial generative nets
(AGNs), a general methodology to train a
generator
neural network to emit adversarial examples satisfying desired objectives. We demonstrate the ability of AGNs to accommodate a wide range of objectives, including imprecise ones difficult to model, in two application domains. In particular, we demonstrate
physical
adversarial examples—eyeglass frames designed to fool face recognition—with better robustness, inconspicuousness, and scalability than previous approaches, as well as a new attack to fool a handwritten-digit classifier.
Funder
NSF
Google and Nvidia, and from Lockheed Martin and NATO through Carnegie Mellon CyLab
CyLab Presidential Fellowship and a Symantec Research Labs Fellowship
National Security Agency
Multidisciplinary University Research Initiative (MURI) Cyber Deception
Publisher
Association for Computing Machinery (ACM)
Subject
Safety, Risk, Reliability and Quality,General Computer Science
Reference79 articles.
1. Anish Athalye and Nicholas Carlini. 2018. On the robustness of the CVPR 2018 white-box adversarial example defenses. arXiv:1804.03286 (2018). Anish Athalye and Nicholas Carlini. 2018. On the robustness of the CVPR 2018 white-box adversarial example defenses. arXiv:1804.03286 (2018).
2. Autodesk. {n.d.}. Measuring light levels. Retrieved from https://goo.gl/hkBWbZ. Autodesk. {n.d.}. Measuring light levels. Retrieved from https://goo.gl/hkBWbZ.
Cited by
98 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献