Quantum machine learning (QML) sits at the intersection of quantum computing and classical machine learning, offering the prospect of new computational paradigms and advantages for processing complex data. This chapter introduces the fundamentals of QML for readers from both communities, establishing a shared conceptual foundation. We connect the worst-case, adversarial perspective from theoretical computer science with the physical principles of quantum systems, highlighting how superposition, entanglement, and measurement collapse influence learning and robustness. Special attention is given to adversarial robustness, understood as the ability of QML models to resist inputs designed to cause failure. We motivate the study of QML in adversarial settings, outlining distinctions between classical and quantum data and computations when the adversary is a core element. This chapter serves as a starting point to adversarial and robust quantum machine learning in subsequent chapters.
Comments
This book chapter has been accepted for Springer Nature Quantum Robustness in Artificial Intelligence and will appear in the book: https://link.springer.com/book/9783032111524#accessibility-information
Quantum machine learning (QML) sits at the intersection of quantum computing and classical machine learning, offering the prospect of new computational paradigms and advantages for processing complex data. This chapter introduces the fundamentals of QML for readers from both communities, establishing a shared conceptual foundation. We connect the worst-case, adversarial perspective from theoretical computer science with the physical principles of quantum systems, highlighting how superposition, entanglement, and measurement collapse influence learning and robustness. Special attention is given to adversarial robustness, understood as the ability of QML models to resist inputs designed to cause failure. We motivate the study of QML in adversarial settings, outlining distinctions between classical and quantum data and computations when the adversary is a core element. This chapter serves as a starting point to adversarial and robust quantum machine learning in subsequent chapters.
This chapter introduces quantum machine learning (QML) as a field at the intersection of quantum computing and machine learning. It presents the fundamental ideas that motivate QML, including the potential of quantum information processing to expand representational power, accelerate selected computational tasks, and support new approaches to learning and inference. This chapter also outlines the major paradigms of QML, including quantum-enhanced, fully quantum, and hybrid quantum-classical models. In addition, this chapter reviews the historical development of the field, the constraints imposed by current noisy intermediate-scale quantum (NISQ) hardware, and the distinction between quantum supremacy and practical utility. It concludes by discussing the long-term path toward fault-tolerant quantum computing and the implications of this roadmap for future quantum learning systems. Together, these topics provide the conceptual foundation for the chapters that follow.
Quantum computing and machine learning represent two of the most promising avenues for breakthroughs in computational science and technology. The chapter explores the foundational aspects and synergies between quantum computing (QC) and machine learning (ML), highlighting how the principles of quantum mechanics can revolutionize data processing and algorithmic efficiency. The basic concepts and architecture of quantum computers, emphasizing their potential to perform calculations exponentially faster than their classical counterparts for specific problems, have been introduced. Following this, the realm of quantum machine learning (QML), a novel field that combines quantum computing with machine learning algorithms to address some of the most challenging data analysis problems, is discussed. The key algorithms and theoretical models that form the basis of QML, including quantum versions of classical algorithms and entirely new quantum approaches to learning, are discussed. Furthermore, the challenges and limitations currently faced by researchers in this field, such as the scarcity of quantum hardware and the complexity of quantum algorithm design, are mentioned. Finally, the future of quantum computing and machine learning, speculating on their potential impacts on various sectors, including cryptography, drug discovery, and artificial intelligence, is explained. Through this exploration, the chapter provides a comprehensive overview of the foundational elements of QC and ML, fostering a deeper understanding of their potential to transform computational methodologies and contribute to the advancement of science and technology.
Abstract Despite their ever more widespread deployment throughout society, machine learning algorithms remain critically vulnerable to being spoofed by subtle adversarial tampering with their input data. The prospect of near-term quantum computers being capable of running quantum machine learning (QML) algorithms has therefore generated intense interest in their adversarial vulnerability. Here we show that quantum properties of QML algorithms can confer fundamental protections against such attacks, in certain scenarios guaranteeing robustness against classically-armed adversaries. We leverage tools from many-body physics to identify the quantum sources of this protection. Our results offer a theoretical underpinning of recent evidence which suggest quantum advantages in the search for adversarial robustness. In particular, we prove that quantum classifiers are: (i) protected against weak perturbations of data drawn from the trained distribution, (ii) protected against local attacks if they are insufficiently scrambling, and (iii) show evidence that they are protected against universal adversarial attacks if they are sufficiently chaotic. Our analytic results are supported by numerical evidence demonstrating the applicability of our theorems and the resulting robustness of a quantum classifier in practice. This line of inquiry constitutes a concrete pathway to advantage in QML, orthogonal to the usually sought improvements in model speed or accuracy.
Quantum Machine Learning (QML) is a new discipline that combines quantum computing and machine learning algorithms in order to take advantage of the computational benefits which are provided by quantum computing for performing complex problems. While similar to classical machine learning systems, quantum-models tend to be prone for adversarial attacks—craft perturbations designed purposely for causing model mispredictions. The intrinsic noise and probabilistic behavior of quantum systems— in particular the (Noisy Intermediate-Scale Quantum) NISQ—is only worsening these vulnerabilities. This paper provides an overview of adversarial threats in QML and illustrates how the quantum nature of any specific system to be attacked shapes both attack and defence strategies. We study on prevalent adversarial methods including gradient based perturbations, optimization attacks and transfer attacks among others, for their resonance in the quantum settings. And, we explore the influence of quantum noise, circuit layout, and hybrid architectures on system weaknesses. In order to develop solutions for such problems, we provide a survey of defines approaches including adversarial training noise-aware learning quantum error mitigation and robust optimization By incorporating probabilistic approaches like Bayesian inference, the ability to make decisions with uncertainty informs and strengthens resilience. Fields like cybersecurity, healthcare and autonomous systems showcase implications of adversarial robustness in QML. Lastly, the paper presents open challenges and future directions for research where some of them are on building scalable secured interpretable quantum AI systems.
This chapter introduces the field of quantum machine learning (QML) by outlining its motivations, foundational concepts, and scope. It provides a high-level overview of how quantum computing principles can be leveraged to enhance or reframe learning tasks, and presents a taxonomy of QML approaches based on the type of data and model used. By bridging classical learning theory with quantum computational models, this chapter sets the stage for deeper exploration of algorithms, architectures, and applications in the chapters that follow.
The integration of quantum computing with machine learning emerged as an approach to solving complex computational problems. This chapter provides an overview of quantum machine learning (QML), exploring the fundamental concepts, algorithms, and applications at the intersection of quantum computing and machine learning. We begin by introducing the basic principles of quantum mechanics relevant to QML, including qubits, quantum gates, and quantum parallelism. Then, we explore key topics such as quantum entanglement, superposition, and their role in quantum algorithms. We then discuss the landscape of QML, contrasting classical and quantum approaches and highlighting the potential of quantum computing to enhance machine learning algorithms. Furthermore, we discuss prominent QML algorithms, such as quantum support vector machines (QSVM) and quantum neural networks (QNN), elucidating their mathematical formulations and differences from classical counterparts. Real-world examples showcasing quantum speedup in machine learning tasks are presented to underscore the transformative potential of quantum computing. Finally, an overview of QML tools and frameworks is provided, enabling researchers to embark on their journey in exploring the exciting field of QML.