Generating Universal Adversarial Perturbations for Quantum Classifiers

Abstract

Quantum Machine Learning (QML) has emerged as a promising field of research, aiming to leverage the capabilities of quantum computing to enhance existing machine learning methodologies. Recent studies have revealed that, like their classical counterparts, QML models based on Parametrized Quantum Circuits (PQCs) are also vulnerable to adversarial attacks. Moreover, the existence of Universal Adversarial Perturbations (UAPs) in the quantum domain has been demonstrated theoretically in the context of quantum classifiers. In this work, we introduce QuGAP: a novel framework for generating UAPs for quantum classifiers. We conceptualize the notion of additive UAPs for PQC-based classifiers and theoretically demonstrate their existence. We then utilize generative models (QuGAP-A) to craft additive UAPs and experimentally show that quantum classifiers are susceptible to such attacks. Moreover, we formulate a new method for generating unitary UAPs (QuGAP-U) using quantum generative models and a novel loss function based on fidelity constraints. We evaluate the performance of the proposed framework and show that our method achieves state-of-the-art misclassification rates, while maintaining high fidelity between legitimate and adversarial samples.

arXiv commentAccepted at AAAI 2024

References in corpus (26)

• Quantum Computing in the NISQ era and beyond
• Explaining and Harnessing Adversarial Examples
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Intriguing properties of neural networks
• Quantum Machine Learning
• Quantum support vector machine for big data classification
• Noisy intermediate-scale quantum (NISQ) algorithms
• Quantum Circuit Learning
• A Survey of Large Language Models
• Quantum principal component analysis
• Evaluating analytic gradients on quantum hardware
• Circuit-centric quantum classifiers
• Delving into Transferable Adversarial Examples and Black-box Attacks
• Synthesis of Quantum Logic Circuits
• Quantum generative adversarial networks
• Robust data encodings for quantum classifiers
• Adversarial Attacks on Deep Learning Models in Natural Language Processing: A Survey
• Quantum noise protects quantum classifiers against adversaries
• Experimental quantum adversarial learning with programmable superconducting qubits
• Vulnerability of quantum classification to adversarial perturbations
• QuGAN: A Quantum State Fidelity based Generative Adversarial Network
• Benchmarking Adversarially Robust Quantum Machine Learning at Scale
• Universal Adversarial Examples and Perturbations for Quantum Classifiers
• projUNN: efficient method for training deep networks with unitary matrices
• Enhancing Quantum Adversarial Robustness by Randomized Encodings
• Certified Robustness of Quantum Classifiers against Adversarial Examples through Quantum Noise

Cited by in corpus (1)

• QuanTest: Entanglement-Guided Testing of Quantum Neural Network Systems