Adversarial defense via learning to generate diverse attacks

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 13
  • Download : 0
With the remarkable success of deep learning, Deep Neural Networks (DNNs) have been applied as dominant tools to various machine learning domains. Despite this success, however, it has been found that DNNs are surprisingly vulnerable to malicious attacks; adding a small, perceptually indistinguishable perturbations to the data can easily degrade classification performance. Adversarial training is an effective defense strategy to train a robust classifier. In this work, we propose to utilize the generator to learn how to create adversarial examples. Unlike the existing approaches that create a one-shot perturbation by a deterministic generator, we propose a recursive and stochastic generator that produces much stronger and diverse perturbations that comprehensively reveal the vulnerability of the target classifier. Our experiment results on MNIST and CIFAR-10 datasets show that the classifier adversarially trained with our method yields more robust performance over various white-box and black-box attacks.
Publisher
Institute of Electrical and Electronics Engineers Inc.
Issue Date
2019-10-29
Language
English
Citation

17th IEEE/CVF International Conference on Computer Vision, ICCV 2019, pp.2740 - 2749

ISSN
1550-5499
DOI
10.1109/ICCV.2019.00283
URI
http://hdl.handle.net/10203/276942
Appears in Collection
CS-Conference Papers(학술회의논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0