DeepAI
Log In Sign Up

Probabilistic Safety for Bayesian Neural Networks

04/21/2020
by   Matthew Wicker, et al.
0

We study probabilistic safety for Bayesian Neural Networks (BNNs) under adversarial input perturbations. Given a compact set of input points, T ⊆ R^m, we study the probability w.r.t. the BNN posterior that all the points in T are mapped to the same, given region S in the output space. In particular, this can be used to evaluate the probability that a network sampled from the BNN is vulnerable to adversarial attacks. We rely on relaxation techniques from non-convex optimization to develop a method for computing a lower bound on probabilistic safety for BNNs, deriving explicit procedures for the case of interval and linear function propagation techniques. We apply our methods to BNNs trained on a regression task, airborne collision avoidance, and MNIST, empirically showing that our approach allows one to certify probabilistic safety of BNNs with thousands of neurons.

READ FULL TEXT
02/10/2021

Bayesian Inference with Certifiable Adversarial Robustness

We consider adversarial training of deep neural networks through the len...
11/05/2020

Sampled Nonlocal Gradients for Stronger Adversarial Attacks

The vulnerability of deep neural networks to small and even imperceptibl...
06/01/2020

Second-Order Provable Defenses against Adversarial Attacks

A robustness certificate is the minimum distance of a given input to the...
04/29/2021

Grid-Free Computation of Probabilistic Safety with Malliavin Calculus

We work with continuous-time, continuous-space stochastic dynamical syst...
09/19/2018

Efficient Formal Safety Analysis of Neural Networks

Neural networks are increasingly deployed in real-world safety-critical ...
10/09/2019

Probabilistic Verification and Reachability Analysis of Neural Networks via Semidefinite Programming

Quantifying the robustness of neural networks or verifying their safety ...
03/08/2020

Adversarial Attacks on Probabilistic Autoregressive Forecasting Models

We develop an effective generation of adversarial attacks on neural mode...