Robustness of Bayesian Neural Networks to White-Box Adversarial Attacks
This addresses the vulnerability of neural networks to adversarial attacks, which is a critical security issue for AI systems, but the approach is incremental as it builds on existing Bayesian and adversarial training methods.
The paper tackled the problem of adversarial robustness in neural networks by investigating Bayesian Neural Networks (BNNs) against white-box attacks, showing that BNNs outperform traditional networks in all experiments and adversarially-trained BNNs often achieve significant improvements over non-Bayesian counterparts.
Bayesian Neural Networks (BNNs), unlike Traditional Neural Networks (TNNs) are robust and adept at handling adversarial attacks by incorporating randomness. This randomness improves the estimation of uncertainty, a feature lacking in TNNs. Thus, we investigate the robustness of BNNs to white-box attacks using multiple Bayesian neural architectures. Furthermore, we create our BNN model, called BNN-DenseNet, by fusing Bayesian inference (i.e., variational Bayes) to the DenseNet architecture, and BDAV, by combining this intervention with adversarial training. Experiments are conducted on the CIFAR-10 and FGVC-Aircraft datasets. We attack our models with strong white-box attacks ($l_\infty$-FGSM, $l_\infty$-PGD, $l_2$-PGD, EOT $l_\infty$-FGSM, and EOT $l_\infty$-PGD). In all experiments, at least one BNN outperforms traditional neural networks during adversarial attack scenarios. An adversarially-trained BNN outperforms its non-Bayesian, adversarially-trained counterpart in most experiments, and often by significant margins. Lastly, we investigate network calibration and find that BNNs do not make overconfident predictions, providing evidence that BNNs are also better at measuring uncertainty.