# The Limitations of Model Uncertainty in Adversarial Settings

arXiv: https://arxiv.org/abs/1812.02606 · HTML: https://mlsanity.com/paper/1812.02606/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Dec 2018.

## Built on by (most cited first, 14 of 14)

- [LiBRe: A Practical Bayesian Approach to Adversarial Detection](https://mlsanity.com/paper/2103.14835/) (2021)
- [A Primer on Bayesian Neural Networks: Review and Debates](https://mlsanity.com/paper/2309.16314/) (2023)
- [Bayesian Inference with Certifiable Adversarial Robustness](https://mlsanity.com/paper/2102.05289/) (2021)
- [On the Robustness of Bayesian Neural Networks to Adversarial Attacks](https://mlsanity.com/paper/2207.06154/) (2022)
- [Adversarial Robustness Guarantees for Gaussian Processes](https://mlsanity.com/paper/2104.03180/) (2021)
- [Exploiting Vulnerabilities in Deep Neural Networks: Adversarial and Fault-Injection Attacks](https://mlsanity.com/paper/2105.03251/) (2021)
- [Adversarial Attacks Against Uncertainty Quantification](https://mlsanity.com/paper/2309.10586/) (2023)
- [The Effect of Prior Lipschitz Continuity on the Adversarial Robustness of Bayesian Neural Networks](https://mlsanity.com/paper/2101.02689/) (2021)
- [Dynamic ensemble selection based on Deep Neural Network Uncertainty Estimation for Adversarial Robustness](https://mlsanity.com/paper/2308.00346/) (2023)
- [Attacking Bayes: On the Adversarial Robustness of Bayesian Neural Networks](https://mlsanity.com/paper/2404.19640/) (2024)
- [A reading survey on adversarial machine learning: Adversarial attacks and their understanding](https://mlsanity.com/paper/2308.03363/) (2023)
- [Adversarial Example Detection in Deployed Tree Ensembles](https://mlsanity.com/paper/2206.13083/) (2022)
- [How Worst-Case Are Adversarial Attacks? Linking Adversarial and Perturbation Robustness](https://mlsanity.com/paper/2601.14519/) (2026)
- [LiST: Lipschitz Scaling Training for Robust and Calibrated Neural Networks](https://mlsanity.com/paper/2607.07745/) (2026)

## Abstract (arXiv)

Machine learning models are vulnerable to adversarial examples: minor perturbations to input samples intended to deliberately cause misclassification. While an obvious security threat, adversarial examples yield as well insights about the applied model itself. We investigate adversarial examples in the context of Bayesian neural network's (BNN's) uncertainty measures. As these measures are highly non-smooth, we use a smooth Gaussian process classifier (GPC) as substitute. We show that both confidence and uncertainty can be unsuspicious even if the output is wrong. Intriguingly, we find subtle differences in the features influencing uncertainty and confidence for most tasks.