To save content items to your account,
please confirm that you agree to abide by our usage policies.
If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account.
Find out more about saving content to .
To save content items to your Kindle, first ensure no-reply@cambridge.org
is added to your Approved Personal Document E-mail List under your Personal Document Settings
on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part
of your Kindle email address below.
Find out more about saving to your Kindle.
Note you can select to save to either the @free.kindle.com or @kindle.com variations.
‘@free.kindle.com’ emails are free but can only be saved to your device when it is connected to wi-fi.
‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.
This book offers a comprehensive introduction to Markov decision process and reinforcement learning fundamentals using common mathematical notation and language. Its goal is to provide a solid foundation that enables readers to engage meaningfully with these rapidly evolving fields. Topics covered include finite and infinite horizon models, partially observable models, value function approximation, simulation-based methods, Monte Carlo methods, and Q-learning. Rigorous mathematical concepts and algorithmic developments are supported by numerous worked examples. As an up-to-date successor to Martin L. Puterman's influential 1994 textbook, this volume assumes familiarity with probability, mathematical notation, and proof techniques. It is ideally suited for students, researchers, and professionals in operations research, computer science, engineering, and economics.
The Cambridge Handbook of Behavioural Data Science offers an essential exploration of how behavioural science and data science converge to study, predict, and explain human, algorithmic, and systemic behaviours. Bringing together scholars from psychology, economics, computer science, engineering, and philosophy, the Handbook presents interdisciplinary perspectives on emerging methods, ethical dilemmas, and real-world applications. Organised into modular parts-Human Behaviour, Algorithmic Behaviour, Systems and Culture, and Applications—it provides readers with a comprehensive, flexible map of the field. Covering topics from cognitive modelling to explainable AI, and from social network analysis to ethics of large language models, the Handbook reflects on both technical innovations and the societal impact of behavioural data, and reinforces concepts in online supplementary materials and videos. The book is an indispensable resource for researchers, students, practitioners, and policymakers who seek to engage critically and constructively with behavioural data in an increasingly digital and algorithmically mediated world.
This volume introduces the fundamental results and the state of the art in infinite duration games on graphs. These versatile game models first appeared in the study of automata and logic, but later became important for program verification and synthesis. They have many more applications. In particular, this book covers models from the fields of optimisation, reinforcement learning, model theory, and set theory. The material is presented at a level accessible to beginning graduate students, requiring only a basic understanding of discrete algorithms and graphs. The book's primary objective is constructing efficient algorithms for analysing different types of games. Rather than describe their implementation in full detail, it investigates their theoretical foundations and proves their properties and winning strategies. This tutorial reference from experts across the field is ideal for students, researchers, and professionals in logic and automata, verification and synthesis, and related fields.
Over the past few decades, graph theory has developed into one of the central areas of modern mathematics, with close (and growing) connections to areas of pure mathematics such as number theory, probability theory, algebra and geometry, as well as to applied areas such as the theory of networks, machine learning, statistical physics, and biology. It is a young and vibrant area, with several major breakthroughs having occurred in just the past few years. This book offers the reader a gentle introduction to the fundamental concepts and techniques of graph theory, covering classical topics such as matchings, colourings and connectivity, alongside the modern and vibrant areas of extremal graph theory, Ramsey theory, and random graphs. The focus throughout is on beautiful questions, ideas and proofs, and on illustrating simple but powerful techniques, such as the probabilistic method, that should be part of every young mathematician's toolkit.
This book presents a modern introduction to the field of algorithmic game theory. It places a heavy emphasis on optimization and online learning (a subdiscipline of machine learning), which are tools that increasingly play a central role in both the theory and practice of applying game-theoretic ideas. The book covers the core techniques used in several majorly successful applications, including techniques used for creating superhuman poker AIs, the theory behind the 'pacing' methodology that has become standard in the internet advertising industry, and the application of competitive equilibrium from equal incomes for fair course seat allocation in many business schools. With its focus on online learning tools, this book is an ideal companion to classic texts on algorithmic game theory for graduate students and researchers.
The burgeoning field of differential equations on graphs has experienced significant growth in the past decade, propelled by the use of variational methods in imaging and by its applications in machine learning. This text provides a detailed overview of the subject, serving as a reference for researchers and as an introduction for graduate students wishing to get up to speed. The authors look through the lens of variational calculus and differential equations, with a particular focus on graph-Laplacian-based models and the graph Ginzburg-Landau functional. They explore the diverse applications, numerical challenges, and theoretical foundations of these models. A meticulously curated bibliography comprising approximately 800 references helps to contextualise this work within the broader academic landscape. While primarily a review, this text also incorporates some original research, extending or refining existing results and methods.
Governing AI is about getting AI right. Building upon AI scholarship in science and technology studies, technology law, business ethics, and computer science, it documents potential risks and actual harms associated with AI, lists proposed solutions to AI-related problems around the world, and assesses their impact. The book presents a vast range of theoretical debates and empirical evidence to document how and how well technical solutions, business self-regulation, and legal regulation work. It is a call to think inside and outside the box. Technical solutions, business self-regulation, and especially legal regulation can mitigate and even eliminate some of the potential risks and actual harms arising from the development and use of AI. However, the long-term health of the relationship between technology and society depends on whether ordinary people are empowered to participate in making informed decisions to govern the future of technology – AI included.
Turing Award-winner Leslie Lamport shares the key lessons he has learned about concurrent and distributed computing over decades of writing and reasoning about their algorithms. Algorithms are not programs, and they shouldn't be written in a programming language. Instead, this book explores how to write them and reason about them by using mathematics. It explains the principles underlying abstract programs, which should be implemented to avoid concurrency errors. Designing an abstract program before writing any code can lead to better, more reliable programs. The book has very few mathematical prerequisites, with an appendix summarizing the necessary knowledge. The examples are written in TLA+ and are available online. This is a fascinating read for any graduate students and researchers in theoretical computer science, concurrency, and distributed systems.
Computable structure theory quantifies and studies the relative complexity of mathematical structures. This text, in conjunction with the author's previous volume, represents the first full monograph on computable structure theory in two decades. It brings new results of the author together with many older results that were previously scattered across the literature and presents them all in a coherent framework. Geared towards graduate students and researchers in mathematical logic, the book enables the reader to learn all the main results and techniques in the area for application in their own research. While the previous volume focused on countable structures whose complexity can be measured within arithmetic, this second volume delves into structures beyond arithmetic, moving into the realm of the hyperarithmetic and the infinitary languages.
For far too long, tech titans peddled promises of disruptive innovation - fabricating benefits and minimizing harms. The promise of quick and easy fixes overpowered a growing chorus of critical voices, driving a sea of private and public investments into increasingly dangerous, misguided, and doomed forms of disruption, with the public paying the price. But what's the alternative? Upgrades - evidence-based, incremental change. Instead of continuing to invest in untested, high-risk innovations, constantly chasing outsized returns, upgraders seek a more proven path to proportional progress. This book dives deep into some of the most disastrous innovations of recent years - the metaverse, cryptocurrency, home surveillance, and AI, to name a few - while highlighting some of the unsung upgraders pushing real progress each day. Timely and corrective, Move Slow and Upgrade pushes us past the baseless promises of innovation, towards realistic hope.
Knowledge-infused learning directly confronts the opacity of current 'black-box' AI models by combining data-driven machine learning techniques with the structured insights of symbolic AI. This guidebook introduces the pioneering techniques of neurosymbolic AI, which blends statistical models with symbolic knowledge to make AI safer and user-explainable. This is critical in high-stakes AI applications in healthcare, law, finance, and crisis management. The book brings readers up to speed on advancements in statistical AI, including transformer models such as BERT and GPT, and provides a comprehensive overview of weakly supervised, distantly supervised, and unsupervised learning methods alongside their knowledge-enhanced variants. Other topics include active learning, zero-shot learning, and model fusion. Beyond theory, the book presents practical considerations and applications of neurosymbolic AI in conversational systems, mental health, crisis management systems, and social and behavioral sciences, making it a pragmatic reference for AI system designers in academia and industry.
In the winter of 2021, the Swedish Nobel Foundation organized a Nobel symposium 'One Hundred Years of Game Theory' to commemorate the publication of famous mathematician Emile Borel's 'La théorie du jeu et les équations intégrales à noyau symétrique'. The symposium gathered roughly forty of the world's most prominent scholars ranging from mathematical foundations to applications in economics, political science, computer science, biology, sociology, and other fields. One Hundred Years of Game Theory brings together their writings to summarize and put in perspective the main achievements of game theory in the last one hundred years. They address past achievements, taking stock of what has been accomplished and contemplating potential future developments and challenges. Offering cross-disciplinary discussions between eminent researchers including five Nobel laureates, one Fields medalist and two Gödel prize winners, the contributors provide a fascinating landscape of game theory and its wide range of applications.
Artificial intelligence is transforming industries and society, but its high energy demands challenge global sustainability goals. Biological intelligence, in contrast, offers both good performance and exceptional energy efficiency. Neuromorphic computing, a growing field inspired by the structure and function of the brain, aims to create energy-efficient algorithms and hardware by integrating insights from biology, physics, computer science, and electrical engineering. This concise and accessible book delves into the principles, mechanisms, and properties of neuromorphic systems. It opens with a primer on biological intelligence, describing learning mechanisms in both simple and complex organisms, then turns to the application of these principles and mechanisms in the development of artificial synapses and neurons, circuits, and architectures. The text also delves into neuromorphic algorithm design, and the unique challenges faced by algorithmic researchers working in this area. The book concludes with a selection of practice problems, with solutions available to instructors online.
This book bridges the gap between theoretical machine learning (ML) and its practical application in industry. It serves as a handbook for shipping production-grade ML systems, addressing challenges often overlooked in academic texts. Drawing on their experience at several major corporations and startups, the authors focus on real-world scenarios, guiding practitioners through the ML lifecycle, from planning and data management to model deployment and optimization. They highlight common pitfalls and offer interview-based case studies from companies that illustrate diverse industrial applications and their unique challenges. Multiple pathways through the book allow readers to choose which stage of the ML development process to focus on, as well as the learning strategy ('crawl,' 'walk,' or 'run') that best suits the needs of their project or team.
Learn to program more effectively, faster, with better results… and enjoy both the learning experience and the benefits it ultimately brings. This undergraduate-level textbook is motivated by Formal Methods, encouraging habits that lead to correct and concise computer programs; but its informal approach sidesteps reliance on Formal Logic that programmers are sometimes led to believe is required. Instead, a straightforward and intuitive use of simple 'What's true here' comments encourages precision of thought without prescription of notation. Drawing on decades of the author's experience in teaching/industry, the text's careful presentation concentrates on key principles of structuring and reasoning about programs, applying them first to small, understandable algorithms. Then students can concentrate on turning those reliably into their corresponding –and correct– program source-codes. The text includes over 200 exercises, with full solutions available online for instructors' use, plus mini-projects and automated quizzes to support instructors in building their own courses.
Aimed at advanced undergraduate and graduate-level students, this textbook covers the core topics of quantum computing in a format designed for a single-semester course. It will be accessible to learners from a range of disciplines, with an understanding of linear algebra being the primary prerequisite. The textbook introduces central concepts such as quantum mechanics, the quantum circuit model, and quantum algorithms, and covers advanced subjects such as the surface code and topological quantum computation. These topics are essential for understanding the role of symmetries in error correction and the stability of quantum architectures, which situate quantum computation within the wider realm of theoretical physics. Graphical representations and exercises are included throughout the book and optional expanded materials are summarized within boxed 'Remarks'. Lecture notes have been made freely available for download from the textbook's webpage, with instructors having additional online access to selected exercise solutions.
What defines a correct program? What education makes a good programmer? The answers to these questions depend on whether programs are seen as mathematical entities, engineered socio-technical systems or media for assisting human thought. Programmers have developed a wide range of concepts and methodologies to construct programs of increasing complexity. This book shows how those concepts and methodologies emerged and developed from the 1940s to the present. It follows several strands in the history of programming and interprets key historical moments as interactions between five different cultures of programming. Rooted in disciplines such as mathematics, electrical engineering, business management or psychology, the different cultures of programming have exchanged ideas and given rise to novel programming concepts and methodologies. They have also clashed about the nature of programming; those clashes remain at the core of many questions about programming today. This title is also available as Open Access on Cambridge Core.
The core topics at the intersection of human-computer interaction (HCI) and US law -- privacy, accessibility, telecommunications, intellectual property, artificial intelligence (AI), dark patterns, human subjects research, and voting -- can be hard to understand without a deep foundation in both law and computing. Every member of the author team of this unique book brings expertise in both law and HCI to provide an in-depth yet understandable treatment of each topic area for professionals, researchers, and graduate students in computing and/or law. Two introductory chapters explaining the core concepts of HCI (for readers with a legal background) and U.S. law (for readers with an HCI background) are followed by in-depth discussions of each topic.