Universal Adversarial Examples and Perturbations for Quantum Classifiers
양자 분류기에 대한 보편적 적대 사례 및 섭동
Ejemplos de adversarios universales y perturbaciones para clasificadores cuánticos
Exemples contradictoires universels et perturbations pour les classificateurs quantiques
Универсальные состязательные примеры и возмущения для квантовых классификаторов
Weiyuan Gong ¹, Dong-Ling Deng 邓东灵 ¹ ²
¹ Center for Quantum Information, IIIS, Tsinghua University, Beijing 100084, People's Republic of China 清华大学 交叉信息研究院 量子信息中心
² Shanghai Qi Zhi Institute, 41th Floor, AI Tower, No. 701 Yunjin Road, Xuhui District, Shanghai 200232, China 上海期智研究院
National Science Review, 22 July 2021

Quantum machine learning explores the interplay between machine learning and quantum physics, which may lead to unprecedented perspectives for both fields. In fact, recent works have shown strong evidences that quantum computers could outperform classical computers in solving certain notable machine learning tasks. Yet, quantum learning systems may also suffer from the vulnerability problem: adding a tiny carefully-crafted perturbation to the legitimate input data would cause the systems to make incorrect predictions at a notably high confidence level.

In this paper, we study the universality of adversarial examples and perturbations for quantum classifiers. Through concrete examples involving classifications of real-life images and quantum phases of matter, we show that there exist universal adversarial examples that can fool a set of different quantum classifiers. We prove that for a set of k classifiers with each receiving input data of n qubits, an O(ln k/2ⁿ) increase of the perturbation strength is enough to ensure a moderate universal adversarial risk.

In addition, for a given quantum classifier we show that there exist universal adversarial perturbations, which can be added to different legitimate samples and make them to be adversarial examples for the classifier.

Our results reveal the universality perspective of adversarial attacks for quantum machine learning systems, which would be crucial for practical applications of both near-term and future quantum technologies in solving machine learning problems.
National Science Review_1
National Science Review_2
National Science Review_3
National Science Review_4
Reviews and Discussions
Soft and Disordered Hyperuniform Elastic Metamaterials for Highly Efficient Vibration Concentration
Accurate and broadband manipulations of harmonic amplitudes and phases to reach 256QAM millimeter-wave wireless communications by time-domain digital coding metasurface
Identification of twist-angle-dependent excitons in WS₂/WSe₂ heterobilayers
H7N9 virus infection triggers lethal cytokine storm by activating gasdermin E-mediated pyroptosis of lung alveolar epithelial cells
The onset of deep recycling of supracrustal materials at the Paleo-Mesoarchean boundary
Water molecules bonded to the carboxylate groups at the inorganic-organic interface of an inorganic nanocrystal coated with alkanoate ligands
Discovery of carbon-based strongest and hardest amorphous material
Cascade-responsive Nanobomb with domino effect for anti-tumor synergistic therapies
Crowded catalyst, better catalyst
Postmortem tissue proteomics reveals the pathogenesis of multiorgan injuries of COVID-19
Wet-chemical synthesis of two-dimensional metal nanomaterials for electrocatalysis
Post-ingestion conversion of dietary indoles into anticancer agents

Previous Article                                Next Article
Copyright © Hot Paper