Adversarial Phenomenon in the Eyes of Bayesian Deep Learning
By Ambrish Rawat, Martin Wistuba, Maria-Irina Nicolae. 2017
42 citations recorded by the source corpus.
Deep Learning models are vulnerable to adversarial examples, i.e.\ images obtained via deliberate imperceptible perturbations, such that the model misclassifies them with high confidence. However, class confidence by itself is an incomplete picture of uncertainty. We therefore use principled Bayesian methods to capture model uncertainty in prediction for observing adversarial misclassification. We provide an extensive study with different Bayesian neural networks attacked in both white-box and black-box setups. The behaviour of the networks for noise, attacks and clean test data is compared. We observe that Bayesian neural networks are uncertain in their predictions for adversarial perturbations, a behaviour similar to the one observed for random Gaussian perturbations. Thus, we conclude that Bayesian neural networks can be considered for detecting adversarial examples.
Subfields: Artificial Intelligence
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Richard S. Zemel
- James Lucas
- Roger Baker Grosse
- Paul Vicol
- Ruslan Salakhutdinov
- Marcel A. J. van Gerven
- Willem F. G. Haselager
- Kuan-Chieh Jackson Wang
Citations in the corpus
- Explanation Methods in Deep Learning: Users, Values, Concerns and Challenges (2018, background): “Work in progress shows that there are methods to detect adversarial instances (Rawat et al., 2017) and to mitigate the attacks (Lin et al., 2017).”
- Adversarial Distillation of Bayesian Neural Network Posteriors (2018, methodology): “Bayesian neural networks are an obvious candidate for this task and thus they have been explored before (Feinman et al., 2017; Louizos & Welling, 2017; Rawat et al., 2017).”
- Continual Learning Using Bayesian Neural Networks (2019, unknown): “There are also some other uncertainty measurements such as computing the entropy of the predictive scores [25] or Model Uncertainty as measured by Mutual Information (MUMMI) [26], see Equation (5).”