Corpus ID: 195767059

Comment on "Adv-BNN: Improved Adversarial Defense through Robust Bayesian Neural Network"

@article{Zimmermann2019CommentO,
  title={Comment on "Adv-BNN: Improved Adversarial Defense through Robust Bayesian Neural Network"},
  author={R. S. Zimmermann},
  journal={ArXiv},
  year={2019},
  volume={abs/1907.00895}
}
  • R. S. Zimmermann
  • Published 2019
  • Mathematics, Computer Science
  • ArXiv
  • A recent paper by Liu et al. combines the topics of adversarial training and Bayesian Neural Networks (BNN) and suggests that adversarially trained BNNs are more robust against adversarial attacks than their non-Bayesian counterparts. Here, I analyze the proposed defense and suggest that one needs to adjust the adversarial attack to incorporate the stochastic nature of a Bayesian network to perform an accurate evaluation of its robustness. Using this new type of attack I show that there appears… CONTINUE READING

    Tables and Topics from this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-4 OF 4 REFERENCES
    Towards Deep Learning Models Resistant to Adversarial Attacks
    • 2,406
    • PDF
    An Analysis of Single-Layer Networks in Unsupervised Feature Learning
    • 1,928
    • PDF
    Synthesizing Robust Adversarial Examples
    • 600
    • PDF
    Adv-BNN: Improved Adversarial Defense through Robust Bayesian Neural Network
    • 46
    • Highly Influential
    • PDF