# Evasion Attacks Against Bayesian Predictive Models

arXiv: https://arxiv.org/abs/2506.09640 · HTML: https://mlsanity.com/paper/2506.09640/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 35 references that are in the map.
- First posted on arXiv 11 Jun 2025.

## Builds on (most cited first, 17 of 17)

- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [Monte Carlo Gradient Estimation in Machine Learning](https://mlsanity.com/paper/1906.10652/) (2019)
- [Manipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learning](https://mlsanity.com/paper/1804.00308/) (2018)
- [Sever: A Robust Meta-Algorithm for Stochastic Optimization](https://mlsanity.com/paper/1803.02815/) (2018)
- [Wild Patterns Reloaded: A Survey of Machine Learning Security against Training Data Poisoning](https://mlsanity.com/paper/2205.01992/) (2022)
- [An Analysis of Adversarial Attacks and Defenses on Autonomous Driving Models](https://mlsanity.com/paper/2002.02175/) (2020)
- [Unbiased MLMC stochastic gradient-based optimization of Bayesian experimental designs](https://mlsanity.com/paper/2005.08414/) (2020)
- [Disrupting Deep Uncertainty Estimation Without Harming Accuracy](https://mlsanity.com/paper/2110.13741/) (2021)
- [Data Poisoning Attacks on Regression Learning and Corresponding Defenses](https://mlsanity.com/paper/2009.07008/) (2020)
- [With Great Dispersion Comes Greater Resilience: Efficient Poisoning Attacks and Defenses for Linear Regression Models](https://mlsanity.com/paper/2006.11928/) (2020)
- [Adversarial Machine Learning: Bayesian Perspectives](https://mlsanity.com/paper/2003.03546/) (2020)
- [Poisoning Bayesian Inference via Data Deletion and Replication](https://mlsanity.com/paper/2503.04480/) (2025)
- [Attacking Bayes: On the Adversarial Robustness of Bayesian Neural Networks](https://mlsanity.com/paper/2404.19640/) (2024)
- [Perturbation Analysis of Learning Algorithms: A Unifying Perspective on Generation of Adversarial Examples](https://mlsanity.com/paper/1812.07385/) (2018)
- [Protecting Classifiers From Attacks](https://mlsanity.com/paper/2004.08705/) (2020)
- [Adversarial Risk Analysis (Overview)](https://mlsanity.com/paper/2007.02613/) (2020)

## Built on by (most cited first, 1 of 1)

- [A probabilistic framework for online test-time adaptation](https://mlsanity.com/paper/2606.26457/) (2026)

## Abstract (arXiv)

There is an increasing interest in analyzing the behavior of machine learning systems against adversarial attacks. However, most of the research in adversarial machine learning has focused on studying weaknesses against evasion or poisoning attacks to predictive models in classical setups, with the susceptibility of Bayesian predictive models to attacks remaining underexplored. This paper introduces a general methodology for designing optimal evasion attacks against such models. We investigate two adversarial objectives: perturbing specific point predictions and altering the entire posterior predictive distribution. For both scenarios, we propose novel gradient-based attacks and study their implementation and properties in various computational setups.