Statistical Guarantees for the Robustness of Bayesian Neural Networks

03/05/2019
by   Luca Cardelli, et al.
0

We introduce a probabilistic robustness measure for Bayesian Neural Networks (BNNs), defined as the probability that, given a test point, there exists a point within a bounded set such that the BNN prediction differs between the two. Such a measure can be used, for instance, to quantify the probability of the existence of adversarial examples. Building on statistical verification techniques for probabilistic models, we develop a framework that allows us to estimate probabilistic robustness for a BNN with statistical guarantees, i.e., with a priori error and confidence bounds. We provide experimental comparison for several approximate BNN inference techniques on image classification tasks associated to MNIST and a two-class subset of the GTSRB dataset. Our results enable quantification of uncertainty of BNN predictions in adversarial settings.

READ FULL TEXT

page 5

page 6

research
09/17/2018

Robustness Guarantees for Bayesian Inference with Gaussian Processes

Bayesian inference and Gaussian processes are widely used in application...
research
05/28/2019

Robustness Quantification for Classification with Gaussian Processes

We consider Bayesian classification with Gaussian processes (GPs) and de...
research
06/23/2023

Adversarial Robustness Certification for Bayesian Neural Networks

We study the problem of certifying the robustness of Bayesian neural net...
research
12/23/2020

Gradient-Free Adversarial Attacks for Bayesian Neural Networks

The existence of adversarial examples underscores the importance of unde...
research
05/29/2018

Lightweight Probabilistic Deep Networks

Even though probabilistic treatments of neural networks have a long hist...
research
05/19/2023

Robust Counterfactual Explanations for Neural Networks With Probabilistic Guarantees

There is an emerging interest in generating robust counterfactual explan...
research
05/11/2022

Robustness Guarantees for Credal Bayesian Networks via Constraint Relaxation over Probabilistic Circuits

In many domains, worst-case guarantees on the performance (e.g., predict...

Please sign up or login with your details

Forgot password? Click here to reset