02/02/2021

Universal Adversarial Perturbations Through the Lens of Deep Steganography: Towards a Fourier Perspective

Chaoning Zhang, Philipp Benz, Adil Karjauv, In So Kweon

Keywords:

Abstract Paper Similar Papers

Abstract: The booming interest in adversarial attacks stems from a misalignment between human vision and a deep neural network (DNN), \ie~a human imperceptible perturbation fools the DNN. Moreover, a single perturbation, often called universal adversarial perturbation (UAP), can be generated to fool the DNN for most images. A similar misalignment phenomenon has also been observed in the deep steganography task, where a decoder network can retrieve a secret image back from a slightly perturbed cover image. We attempt explaining the success of both in a unified manner from the Fourier perspective. We perform task-specific and joint analysis and reveal that (a) frequency is a key factor that influences their performance based on the proposed entropy metric for quantifying the frequency distribution; (b) their success can be attributed to a DNN being highly sensitive to high-frequency content. We also perform feature layer analysis for providing deep insight on model generalization and robustness. Additionally, we propose two new variants of universal perturbations: (1) high-pass UAP (HP-UAP) being less visible to the human eye; (2) Universal Secret Adversarial Perturbation (USAP) that simultaneously achieves attack and hiding.

The video of this talk cannot be embedded. You can watch it here:
https://slideslive.com/38949090
(Link will open in new window)
 0
 0
 0
 0
  Share    
This is an embedded video. Talk and the respective paper are published at AAAI 2021 virtual conference. If you are one of the authors of the paper and want to manage your upload, see the question "My papertalk has been externally embedded..." in the FAQ section.

Comments

Post Comment
no comments yet
code of conduct: tbd Characters remaining: 140

Similar Papers

 0
 0
 0
 0
 3:23 
 0
 0
 0
 0
 18:59 
 0
 0
 0
 0
 12:35 
 0
 0
 0
 0
 3:04 
 0
 0
 0
 0
 17:54 
 0
 0
 0
 0
 1:01 
 0
 0
 0
 0
 1:01 
 0
 0
 0
 0
 9:28 
 0
 0
 0
 0
 14:19 
 0
 0
 0
 0
 15:15 
 0
 0
 0
 0
 5:20 
 0
 0
 0
 0
 17:14 
 0
 0
 0
 0
 3:21 
 0
 0
 0
 0
 9:34 
 0
 0
 0
 0
 12:11 
 0
 0
 0
 0
 11:27 
 0
 0
 0
 0
 1:00 
 0
 0
 0
 0
 13:36 
 0
 0
 0
 0
 1:01 
 0
 0
 0
 0
 14:46 
 0
 0
 0
 0
 12:48 
 0
 0
 0
 0
 5:22 
 0
 0
 0
 0
 3:59 
 0
 0
 0
 0
 15:06 
 0
 0
 0
 0
 12:27 
 0
 0
 0
 0
 3:51 
 0
 0
 0
 0
 15:14 
 0
 0
 0
 0
 3:12 
 0
 0
 0
 0
 10:28 
 0
 0
 0
 0
 17:17 
 0
 0
 0
 0
 4:52 
 0
 0
 0
 0
 2:57 
 0
 0
 0
 0
 1:01 
 0
 0
 0
 0
 1:01 
 0
 0
 0
 0
 12:37 
 0
 0
 0
 0
 14:47 
 0
 0
 0
 0
 3:59 
 0
 0
 0
 0
 14:50 
 0
 0
 0
 0
 4:53 
 0
 0
 0
 0
 5:08