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

2026-02-14

AI Summary

This work addresses the high computational complexity of verifying robustness in binary neural networks (BNNs) under input perturbations, including adversarial attacks. It is the first to systematically formulate this verification task as a Quadratically Constrained Boolean Optimization (QCBO) problem and subsequently transform it into a Quadratic Unconstrained Binary Optimization (QUBO) form, thereby enabling compatibility with Ising models and quantum-inspired solvers. The resulting formulation is efficiently solved using free-energy machines, simulated annealing, and quantum/digital annealing platforms, effectively bridging classical robustness verification with emerging Ising-based computational paradigms. Experimental validation on binarized MNIST demonstrates the feasibility of the proposed framework, offering a novel pathway toward trustworthy AI through the integration of quantum-inspired computing.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Constraint Satisfaction and Optimization: Satisfiability

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

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.

Problem

• Binary Neural Networks
• Robustness Verification
• Adversarial Attacks
• Combinatorial Optimization
• Input Perturbations

Innovation

• Binary Neural Networks
• Robustness Verification
• QUBO
• Ising Model
• Quantum-Inspired Computing