Towards Deep Learning Models Resistant to Adversarial Attacks

@article{Madry2018TowardsDL,
  title={Towards Deep Learning Models Resistant to Adversarial Attacks},
  author={Aleksander Madry and Aleksandar Makelov and Ludwig Schmidt and Dimitris Tsipras and Adrian Vladu},
  journal={CoRR},
  year={2018},
  volume={abs/1706.06083}
}
Recent work has demonstrated that neural networks are vulnerable to adversarial examples, i.e., inputs that are almost indistinguishable from natural data and yet classified incorrectly by the network. In fact, some of the latest findings suggest that the existence of adversarial attacks may be an inherent weakness of deep learning models. To address this problem, we study the adversarial robustness of neural networks through the lens of robust optimization. This approach provides us with a… CONTINUE READING

Citations

Publications citing this paper.
SHOWING 1-10 OF 592 CITATIONS, ESTIMATED 39% COVERAGE

Rethinking Softmax Cross-Entropy Loss for Adversarial Robustness

  • ArXiv
  • 2019
VIEW 22 EXCERPTS
CITES BACKGROUND, METHODS & RESULTS
HIGHLY INFLUENCED

Adversarial Defense by Restricting the Hidden Space of Deep Neural Networks

VIEW 22 EXCERPTS
CITES BACKGROUND, RESULTS & METHODS
HIGHLY INFLUENCED

Adversarial Training with Voronoi Constraints

VIEW 20 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Adversarially Robust Generalization Just Requires More Unlabeled Data

Runtian Zhai, Tianle Cai, +4 authors Liwei Wang
  • ArXiv
  • 2019
VIEW 21 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Adversarially Trained Model Compression: When Robustness Meets Efficiency

  • ArXiv
  • 2019
VIEW 6 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Bamboo: Ball-Shape Data Augmentation Against Adversarial Attacks from All Directions

  • SafeAI@AAAI
  • 2019
VIEW 8 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Beyond Adversarial Training: Min-Max Optimization in Adversarial Attack and Defense

Jingkang Wang, Tianyun Zhang, +4 authors Bo Li
  • ArXiv
  • 2019
VIEW 10 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Boundary Attack++: Query-Efficient Decision-Based Adversarial Attack

VIEW 12 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

FILTER CITATIONS BY YEAR

2016
2019

CITATION STATISTICS

  • 247 Highly Influenced Citations

  • Averaged 116 Citations per year over the last 3 years

  • 89% Increase in citations per year in 2018 over 2017

References

Publications referenced by this paper.

Similar Papers

Loading similar papers…