The Effect of Prior Lipschitz Continuity on the Adversarial Robustness of Bayesian Neural Networks

Abstract

It is desirable, and often a necessity, for machine learning models to be robust against adversarial attacks. This is particularly true for Bayesian models, as they are well-suited for safety-critical applications, in which adversarial attacks can have catastrophic outcomes. In this work, we take a deeper look at the adversarial robustness of Bayesian Neural Networks (BNNs). In particular, we consider whether the adversarial robustness of a BNN can be increased by model choices, particularly the Lipschitz continuity induced by the prior. Conducting in-depth analysis on the case of i.i.d., zero-mean Gaussian priors and posteriors approximated via mean-field variational inference, we find evidence that adversarial robustness is indeed sensitive to the prior variance.

arXiv comment4 pages, 2 tables, AAAI 2021 Workshop Towards Robust, Secure and Efficient Machine Learning

References in corpus (16)

• Towards Deep Learning Models Resistant to Adversarial Attacks
• Wild Patterns: Ten Years After the Rise of Adversarial Machine Learning
• Weight Uncertainty in Neural Networks
• Detecting Adversarial Samples from Artifacts
• Bayesian Deep Learning and a Probabilistic Perspective of Generalization
• Understanding Measures of Uncertainty for Adversarial Example Detection
• Adversarial Attacks Against Medical Deep Learning Systems
• Adv-BNN: Improved Adversarial Defense through Robust Bayesian Neural Network
• Bayesian Learning-Based Adaptive Control for Safety Critical Systems
• A Closer Look at Accuracy vs. Robustness
• ME-Net: Towards Effective Adversarial Robustness with Matrix Estimation
• How Good is the Bayes Posterior in Deep Neural Networks Really?
• Robustness of Bayesian Neural Networks to Gradient-Based Attacks
• The Limitations of Model Uncertainty in Adversarial Settings
• Intriguing Properties of Adversarial Examples
• Comment on "Adv-BNN: Improved Adversarial Defense through Robust Bayesian Neural Network"