Verifying Properties of Binarized Deep Neural Networks

TL;DR

This paper introduces a novel exact Boolean encoding of Binarized Neural Networks, enabling the use of SAT solvers to verify properties like robustness to adversarial attacks in deep learning models.

Contribution

It presents the first exact Boolean representation of deep neural networks, facilitating property verification using SAT solvers and a counterexample-guided search.

Findings

Scales to medium-sized image classification networks

Enables verification of robustness to adversarial perturbations

First exact Boolean encoding for deep neural network verification

Abstract

Understanding properties of deep neural networks is an important challenge in deep learning. In this paper, we take a step in this direction by proposing a rigorous way of verifying properties of a popular class of neural networks, Binarized Neural Networks, using the well-developed means of Boolean satisfiability. Our main contribution is a construction that creates a representation of a binarized neural network as a Boolean formula. Our encoding is the first exact Boolean representation of a deep neural network. Using this encoding, we leverage the power of modern SAT solvers along with a proposed counterexample-guided search procedure to verify various properties of these networks. A particular focus will be on the critical property of robustness to adversarial perturbations. For this property, our experimental results demonstrate that our approach scales to medium-size deep neural networks used in image classification tasks. To the best of our knowledge, this is the first work on verifying properties of deep neural networks using an exact Boolean encoding of the network.