Low Frequency Adversarial Perturbation Low Frequency Adversarial Perturbation
Paper summary Guo et al. propose to augment black-box adversarial attacks with low-frequency noise to obtain low-frequency adversarial examples as shown in Figure 1. To this end, the boundary attack as well as the NES attack are modified to sample from a low-frequency Gaussian distribution instead from Gaussian noise directly. This is achieved through an inverse discrete cosine transform as detailed in the paper. https://i.imgur.com/fejvuw7.jpg Figure 1: Example of a low-frequency adversarial example. Also find this summary at [davidstutz.de](https://davidstutz.de/category/reading/).
Low Frequency Adversarial Perturbation
Chuan Guo and Jared S. Frank and Kilian Q. Weinberger
arXiv e-Print archive - 2018 via Local arXiv
Keywords: cs.CV


Summary by David Stutz 1 year ago
Your comment:

ShortScience.org allows researchers to publish paper summaries that are voted on and ranked!

Sponsored by: and