Robustness Verification of Binary Neural Networks: An Ising and Quantum-Inspired Framework

TL;DR

This paper introduces an Ising- and quantum-inspired framework for verifying the robustness of binary neural networks against input perturbations, leveraging quantum-inspired solvers and Ising computing to address computational challenges.

Contribution

It formulates BNN robustness verification as a QCBO problem and transforms it into a QUBO instance suitable for Ising and quantum-inspired solvers, demonstrating practical feasibility.

Findings

Successfully verified BNN robustness on binarized MNIST using QUBO solvers.

Demonstrated deployment on quantum annealing and digital annealing platforms.

Showed potential of quantum-inspired computing for trustworthy AI systems.

Abstract

Binary neural networks (BNNs) are increasingly deployed in edge computing applications due to their low hardware complexity and high energy efficiency. However, verifying the robustness of BNNs against input perturbations, including adversarial attacks, remains computationally challenging because the underlying decision problem is inherently combinatorial. In this paper, we propose an Ising- and quantum-inspired framework for BNN robustness verification. We show that, for a broad class of BNN architectures, robustness verification can be formulated as a Quadratic Constrained Boolean Optimization (QCBO) problem and subsequently transformed into a Quadratic Unconstrained Boolean Optimization (QUBO) instance amenable to Ising and quantum-inspired solvers. We demonstrate the feasibility of this formulation on binarized MNIST by solving the resulting QUBOs with a free energy machine (FEM) solver and simulated annealing. We also show the deployment of this framework on quantum annealing and digital annealing platforms. Our results highlight the potential of quantum-inspired computing and Ising computing as a pathway toward trustworthy AI systems.