Comment on "Adv-BNN: Improved Adversarial Defense through Robust Bayesian Neural Network"
Not on the map. Why?
Abstract (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 to be no strong evidence for higher robustness of the adversarially trained BNNs.
Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). First posted on arXiv 1 Jul 2019.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 3; 2022: 2; 2023: 2; 2024: 3; 2025: 1; 2026: 1.
12 papers.
- 2021GradDiv: Adversarial Robustness of Randomized Neural Networks via Gradient Diversity Regularizationcited by 34
- 2021Benchmarking the Accuracy and Robustness of Feedback Alignment Algorithmscited by 9
- 2024LoRID: Low-Rank Iterative Diffusion for Adversarial Purificationcited by 7
- 2022Bayesian Learning with Information Gain Provably Bounds Risk for a Robust Adversarial Defensecited by 4
- 2021The Effect of Prior Lipschitz Continuity on the Adversarial Robustness of Bayesian Neural Networkscited by 3
- 2022Mitigating Closed-model Adversarial Examples with Bayesian Neural Modeling for Enhanced End-to-End Speech Recognitioncited by 2
- 2024Attacking Bayes: On the Adversarial Robustness of Bayesian Neural Networkscited by 2
- 2024Improving Adversarial Robustness via Decoupled Visual Representation Maskingcited by 2
- 2023Spatial-Frequency Discriminability for Revealing Adversarial Perturbationscited by 1
- 2023Transgressing the boundaries: towards a rigorous understanding of deep learning and its (non-)robustnesscited by 1
- 2025How to Backdoor Image Knowledge Distillationcited by 1
- 2026A combination of noise and bilateral filters achieve supralinear and scalable adversarial robustness in CNNscited by 0
- 2026A combination of noise and bilateral filters achieve supralinear and scalable adversarial robustness in CNNs1 Jun 2026
- 2025How to Backdoor Image Knowledge Distillation30 Apr 2025
- 2024LoRID: Low-Rank Iterative Diffusion for Adversarial Purification12 Sep 2024
- 2024Improving Adversarial Robustness via Decoupled Visual Representation Masking16 Jun 2024
- 2024Attacking Bayes: On the Adversarial Robustness of Bayesian Neural Networks27 Apr 2024
- 2023Transgressing the boundaries: towards a rigorous understanding of deep learning and its (non-)robustness5 Jul 2023
- 2023Spatial-Frequency Discriminability for Revealing Adversarial Perturbations18 May 2023
- 2022Bayesian Learning with Information Gain Provably Bounds Risk for a Robust Adversarial Defense5 Dec 2022
- 2022Mitigating Closed-model Adversarial Examples with Bayesian Neural Modeling for Enhanced End-to-End Speech Recognition17 Feb 2022
- 2021Benchmarking the Accuracy and Robustness of Feedback Alignment Algorithms30 Aug 2021
- 2021GradDiv: Adversarial Robustness of Randomized Neural Networks via Gradient Diversity Regularization6 Jul 2021
- 2021The Effect of Prior Lipschitz Continuity on the Adversarial Robustness of Bayesian Neural Networks7 Jan 2021