Search

Papers from a flagship conference reflect the latest developments in the field, including work in such rapidly advancing areas as human-robot interaction and formal methods.Robotics: Science and Systems VII spans a wide spectrum of robotics, bringing together researchers working on the algorithmic or mathematical foundations of robotics, robotics applications, and analysis of robotics systems. This volume presents the proceedings of the seventh annual Robotics: Science and Systems conference, held in 2011 at the University of Southern California. The papers presented cover a wide range of topics in robotics, spanning mechanisms, kinematics, dynamics and control, human-robot interaction and human-centered systems, distributed systems, mobile systems and mobility, manipulation, field robotics, medical robotics, biological robotics, robot perception, and estimation and learning in robotic systems. The conference and its proceedings reflect not only the tremendous growth of robotics as a discipline but also the desire in the robotics community for a flagship event at which the best of the research in the field can be presented.

Methods by which robots can learn control laws that enable real-time reactivity using dynamical systems; with applications and exercises.This book presents a wealth of machine learning techniques to make the control of robots more flexible and safe when interacting with humans. It introduces a set of control laws that enable reactivity using dynamical systems, a widely used method for solving motion-planning problems in robotics. These control approaches can replan in milliseconds to adapt to new environmental constraints and offer safe and compliant control of forces in contact. The techniques offer theoretical advantages, including convergence to a goal, non-penetration of obstacles, and passivity. The coverage of learning begins with low-level control parameters and progresses to higher-level competencies composed of combinations of skills. Learning for Adaptive and Reactive Robot Control is designed for graduate-level courses in robotics, with chapters that proceed from fundamentals to more advanced content. Techniques covered include learning from demonstration, optimization, and reinforcement learning, and using dynamical systems in learning control laws, trajectory planning, and methods for compliant and force control. Features for teaching in each chapter: applications, which range from arm manipulators to whole-body control of humanoid robots;pencil-and-paper and programming exercises;lecture videos, slides, and MATLAB code examples available on the author's website. an eTextbook platform website offering protected material[EPS2] for instructors including solutions.

A rigorous and comprehensive textbook covering the major approaches to knowledge graphs, an active and interdisciplinary area within artificial intelligence.The field of knowledge graphs, which allows us to model, process, and derive insights from complex real-world data, has emerged as an active and interdisciplinary area of artificial intelligence over the last decade, drawing on such fields as natural language processing, data mining, and the semantic web. Current projects involve predicting cyberattacks, recommending products, and even gleaning insights from thousands of papers on COVID-19. This textbook offers rigorous and comprehensive coverage of the field. It focuses systematically on the major approaches, both those that have stood the test of time and the latest deep learning methods.

An accessible introduction to algorithms, explaining not just what they are but how they work, with examples from a wide range of application areas.Digital technology runs on algorithms, sets of instructions that describe how to do something efficiently. Application areas range from search engines to tournament scheduling, DNA sequencing, and machine learning. Arguing that every educated person today needs to have some understanding of algorithms and what they do, in this volume in the MIT Press Essential Knowledge series, Panos Louridas offers an introduction to algorithms that is accessible to the nonspecialist reader. Louridas explains not just what algorithms are but also how they work, offering a wide range of examples and keeping mathematics to a minimum.After discussing what an algorithm does and how its effectiveness can be measured, Louridas covers three of the most fundamental applications areas: graphs, which describe networks, from eighteenth-century problems to today's social networks; searching, and how to find the fastest way to search; and sorting, and the importance of choosing the best algorithm for particular tasks. He then presents larger-scale applications: PageRank, Google's founding algorithm; and neural networks and deep learning. Finally, Louridas describes how all algorithms are nothing more than simple moves with pen and paper, and how from such a humble foundation rise all their spectacular achievements.

Proceedings from the third annual Robotics: Science and Systems conference, presenting state-of-the-art research on the foundations of robotics, robotics applications, and robotics systems.Robotics: Science and Systems III spans a wide spectrum of robotics, bringing together researchers working on the foundations of robotics, robotics applications, and analysis of robotics systems. This volume presents the proceedings of the third annual Robotics: Science and Systems conference, held in June 2007 at Georgia Tech. Papers report state-of-the-art research on topics as diverse as Legged Robotics, Reconfigurable Robots, Biomimetic Robots, Manipulation, Humanoid Robotics, Telerobotics, Haptics, Motion Planning, Collision Avoidance, Robot Vision and Perception, Bayesian Techniques, Machine Learning, Mobile Robots, and Multi-robot systems. This conference reflects not only the tremendous growth of robotics as a discipline but also the desire in the robotics community for a flagship event at which the best of the research in the field can be presented.

A timely, practical guide to AI—its strengths, weaknesses, and real-world applications—for business professionals and policymakers.Artificial intelligence, or AI, can recognize a pattern from any set of data it is given, which is what makes it such an extraordinarily powerful tool. But because not all patterns are authentic or reliable, AI's pattern-finding superpower can lead to spurious patterns—and to disastrous results for business and government entities that rely on them. Hence the conundrum at the heart of AI: its greatest strength can also be its greatest weakness. Targeting the businessperson who needs to know how to use AI profitably and responsibly, Caleb Briggs and Rex Briggs offer in this book a foundational understanding of AI that is easy to grasp yet thorough enough to be used effectively. The AI Conundrum:• Draws on the authors'diverse expertise—in pure math, computer science, marketing, data science, and business—to make AI concepts and applications approachable for readers of all tech levels.• Provides a framework for comparing AI to the next best alternative, and for gauging where AI is likely be successful—or to pose greater risk than benefits.• Includes dozens of real-world case studies highlighting the successes and failures of AI applications across various industries. • Offers actionable insights for responsible implementation and risk mitigation.• Provides a worksheet for identifying potential problem areas, a cost-benefit analysis, and a companion website. The AI Conundrum is an invaluable resource for professionals and students seeking a full understanding of AI—its applications, limitations, and ethical considerations—as we enter a brave new era.

The first text to take a systems engineering approach to artificial intelligence (AI), from architecture principles to the development and deployment of AI capabilities.Most books on artificial intelligence (AI) focus on a single functional building block, such as machine learning or human-machine teaming. Artificial Intelligence takes a more holistic approach, addressing AI from the view of systems engineering. The book centers on the people-process-technology triad that is critical to successful development of AI products and services. Development starts with an AI design, based on the AI system architecture, and culminates with successful deployment of the AI capabilities. Directed toward AI developers and operational users, this accessibly written volume of the MIT Lincoln Laboratory Series can also serve as a text for undergraduate seniors and graduate-level students and as a reference book. Key features:In-depth look at modern computing technologies Systems engineering description and means to successfully undertake an AI product or service development through deploymentExisting methods for applying machine learning operations (MLOps)AI system architecture including a description of each of the AI pipeline building blocksChallenges and approaches to attend to responsible AI in practice Tools to develop a strategic roadmap and techniques to foster an innovative team environment Multiple use cases that stem from the authors'MIT classes, as well as from AI practitioners, AI project managers, early-career AI team leaders, technical executives, and entrepreneurs Exercises and Jupyter notebook examples

A deep look into the multifaceted landscape of artificial intelligence, considering AI's ethical and societal implications and celebrating its diverse and innovative applications.In this edited collection Collaborative Intelligence, Mira Lane and Arathi Sethumadhavan embark on a captivating exploration of artificial intelligence. The book delves deeply into the dynamic interplay between theory and practice, shedding light on the transformative potential and complexities of AI. For practitioners deeply immersed in the world of AI, Lane and Sethumadhavan offer firsthand accounts and insights from technologists, academics, and thought leaders, as well as a series of compelling case studies, ranging from AI's impact on artistry to its role in addressing societal challenges like modern slavery and wildlife conservation. As the global AI market burgeons, this book enables collaboration, knowledge sharing, and interdisciplinary dialogue. It caters not only to the practitioners shaping the AI landscape but also to policymakers striving to navigate the intricate relationship between humans and machines, as well as academics. Divided into two parts, the first half of the book offers readers a comprehensive understanding of AI's historical context, its influence on power dynamics, human-AI interaction, and the critical role of audits in governing AI systems. The second half unfolds a series of eight case studies, unraveling AI's impact on fields as varied as healthcare, vehicular safety, conservation, human rights, and the metaverse. Each chapter in this book paints a vivid picture of AI's triumphs and challenges, providing a panoramic view of how it is reshaping our world.

A concise and illuminating history of the field of artificial intelligence from one of its earliest and most respected pioneers.AI & I is an intellectual history of the field of artificial intelligence from the perspective of one of its first practitioners, Eugene Charniak. Charniak entered the field in 1967, roughly 12 years after AI's founding, and was involved in many of AI's formative milestones. In this book, he traces the trajectory of breakthroughs and disappointments of the discipline up to the current day, clearly and engagingly demystifying this oft revered and misunderstood technology. His argument is controversial but well supported: that classical AI has been almost uniformly unsuccessful and that the modern deep learning approach should be viewed as the foundation for all the exciting developments that are to come.Written for the scientifically educated layperson, this book chronicles the history of the field of AI, starting with its origin in 1956, as a topic for a small academic workshop held at Dartmouth University. From there, the author covers reasoning and knowledge representation, reasoning under uncertainty, chess, computer vision, speech recognition, language acquisition, deep learning, and learning writ large. Ultimately, Charniak takes issue with the controversy of AI—the fear that its invention means the end of jobs, creativity, and potentially even humans as a species—and explains why such concerns are unfounded. Instead, he believes that we should embrace the technology and all its potential to benefit society.

An anthology of seven decades of English-language outputs from computer generation systems, chronicling the vast history of machine-written texts created long before ChatGPT.The discussion of computer-generated text has recently reached a fever pitch but largely omits the long history of work in this area—text generation, as it happens, was not invented yesterday in Silicon Valley. This anthology, Output, thoughtfully selected, introduced, and edited by Lillian-Yvonne Bertram and Nick Montfort, aims to correct that omission by gathering seven decades of English-language texts produced by generation systems and software. The outputs span many different types of creative writing and include text generated by research systems, along with reports and utilitarian texts, representing many general advances and experiments in text generation.Output is first and foremost a collection of outputs to be encountered by readers. In addition to an overall introduction, each of the excerpts is introduced individually and organized by fine-grain genre including conversations, humor, letters, poetry, prose, and sentences. Bibliographic references allow readers to learn more about outputs and systems that intrigue them. Although Output could serve as a reference book, it is designed to be readable and to be read. Purposefully excluded are human–computer collaborations that were conceptually defined but not implemented as a computer system.Copublished by Counterpath Press

How AI will challenge our ideas about personhood.Chatbots like ChatGPT have challenged human exceptionalism: we are no longer the only beings capable of generating language and ideas fluently. But is ChatGPT conscious? Or is it merely engaging in sophisticated mimicry? And what happens in the future if the claims to consciousness are more credible? In The Line, James Boyle explores what these changes might do to our concept of personhood, to “the line” we believe separates our species from the rest of the world but that also separates “persons” with legal rights from objects.The personhood wars—over the rights of corporations and animals, over the question of when life begins and ends—have always been contentious. We've even denied the personhood of members of our own species. How will those old fights affect the new ones, and vice versa? Boyle pursues these questions across a dizzying array of fields. He discusses moral philosophy and science fiction, transgenic species, nonhuman animals, the surprising history of corporate personality, and AI itself. Engaging with empathy and anthropomorphism, courtroom battles on behalf of chimps, and doom-laden projections about the threat of AI, The Line offers fascinating and thoughtful answers to questions about our future that will arrive sooner than we think.

A novel approach to hybrid AI aimed at developing trustworthy agent collaborators.The vast majority of current AI relies wholly on machine learning (ML). However, the past thirty years of effort in this paradigm have shown that, despite the many things that ML can achieve, it is not an all-purpose solution to building human-like intelligent systems. One hope for overcoming this limitation is hybrid AI: that is, AI that combines ML with knowledge-based processing. In Agents in the Long Game of AI, Marjorie McShane, Sergei Nirenburg, and Jesse English present recent advances in hybrid AI with special emphases on content-centric computational cognitive modeling, explainability, and development methodologies. At present, hybridization typically involves sprinkling knowledge into an ML black box. The authors, by contrast, argue that hybridization will be best achieved in the opposite way: by building agents within a cognitive architecture and then integrating judiciously selected ML results. This approach leverages the power of ML without sacrificing the kind of explainability that will foster society's trust in AI. This book shows how we can develop trustworthy agent collaborators of a type not being addressed by the “ML alone” or “ML sprinkled by knowledge” paradigms—and why it is imperative to do so.

How to make AI capable of general intelligence, and what such technology would mean for society.Artificial intelligence surrounds us. More and more of the systems and services you interact with every day are based on AI technology. Although some very recent AI systems are generalists to a degree, most AI is narrowly specific; that is, it can only do a single thing, in a single context. For example, your spellchecker can't do mathematics, and the world's best chess-playing program can't play Tetris. Human intelligence is different. We can solve a variety of tasks, including those we have not seen before. In Artificial General Intelligence, Julian Togelius explores technical approaches to developing more general artificial intelligence and asks what general AI would mean for human civilization.Togelius starts by giving examples of narrow AI that have superhuman performance in some way. Interestingly, there have been AI systems that are superhuman in some sense for more than half a century. He then discusses what it would mean to have general intelligence, by looking at definitions from psychology, ethology, and computer science. Next, he explores the two main families of technical approaches to developing more general artificial intelligence: foundation models through self-supervised learning, and open-ended learning in virtual environments. The final chapters of the book investigate potential artificial general intelligence beyond the strictly technical aspects. The questions discussed here investigate whether such general AI would be conscious, whether it would pose a risk to humanity, and how it might alter society.

An accessible, authoritative, and up-to-date computer vision textbook offering a comprehensive introduction to the foundations of the field that incorporates the latest deep learning advances.Machine learning has revolutionized computer vision, but the methods of today have deep roots in the history of the field. Providing a much-needed modern treatment, this accessible and up-to-date textbook comprehensively introduces the foundations of computer vision while incorporating the latest deep learning advances. Taking a holistic approach that goes beyond machine learning, it addresses fundamental issues in the task of vision and the relationship of machine vision to human perception. Foundations of Computer Vision covers topics not standard in other texts, including transformers, diffusion models, statistical image models, issues of fairness and ethics, and the research process. To emphasize intuitive learning, concepts are presented in short, lucid chapters alongside extensive illustrations, questions, and examples. Written by leaders in the field and honed by a decade of classroom experience, this engaging and highly teachable book offers an essential next-generation view of computer vision. Up-to-date treatment integrates classic computer vision and deep learning Accessible approach emphasizes fundamentals and assumes little background knowledgeStudent-friendly presentation features extensive examples and imagesProven in the classroomInstructor resources include slides, solutions, and source code

How Big Tech is taking advantage of us, how AI is making it worse, and how we can create a thriving, AI-positive world.On balance, will AI help humanity or harm it? AI could revolutionize science, medicine, and technology, and deliver us a world of abundance and better health. Or it could be a disaster, leading to the downfall of democracy, or even our extinction. In Taming Silicon Valley, Gary Marcus, one of the most trusted voices in AI, explains that we still have a choice. And that the decisions we make now about AI will shape our next century. In this short but powerful manifesto, Marcus explains how Big Tech is taking advantage of us, how AI could make things much worse, and, most importantly, what we can do to safeguard our democracy, our society, and our future.Marcus explains the potential—and potential risks—of AI in the clearest possible terms and how Big Tech has effectively captured policymakers. He begins by laying out what is lacking in current AI, what the greatest risks of AI are, and how Big Tech has been playing both the public and the government, before digging into why the US government has thus far been ineffective at reining in Big Tech. He then offers real tools for readers, including eight suggestions for what a coherent AI policy should look like—from data rights to layered AI oversight to meaningful tax reform—and closes with how ordinary citizens can push for what is so desperately needed.Taming Silicon Valley is both a primer on how AI has gotten to its problematic present state and a book of activism in the tradition of Abbie Hoffman's Steal This Book and Thomas Paine's Common Sense. It is a deeply important book for our perilous historical moment that every concerned citizen must read.

A fresh look at a “robot-proof” education in the new age of generative AI.In 2017, Robot-Proof, the first edition, foresaw the advent of the AI economy and called for a new model of higher education designed to help human beings flourish alongside smart machines. That economy has arrived. Creative tasks that, seven years ago, seemed resistant to automation can now be performed with a simple prompt. As a result, we must now learn not only to be conversant with these technologies, but also to comprehend and deploy their outputs. In this revised and updated edition, Joseph Aoun rethinks the university's mission for a world transformed by AI, advocating for the lifelong endeavor of a “robot-proof” education. Aoun puts forth a framework for a new curriculum, humanics, which integrates technological, data, and human literacies in an experiential setting, and he renews the call for universities to embrace lifelong learning through a social compact with government, employers, and learners themselves. Drawing on the latest developments and debates around generative AI, Robot-Proof is a blueprint for the university as a force for human reinvention in an era of technological change—an era in which we must constantly renegotiate the shifting boundaries between artificial intelligence and the capacities that remain uniquely human.

How AI can positively impact so many aspects of our daily lives, from health and wellness to work, education, and home life.Artificial intelligence (AI) is a powerful general-purpose technology that is reshaping the modern economy, but misperceptions about AI stand in the way of harnessing it for the betterment of humanity. In Thrive, Ravi Bapna and Anindya Ghose counter the backlash by showcasing how AI is positively influencing the aspects of our daily lives that we care about most: our health and wellness, relationships, education, the workplace, and domestic life. In the process the authors help explain the underlying technology and give people the agency they need to shape the debate around how we should regulate AI to maximize its benefits and minimize its risks.Bringing over two decades of experience with cutting-edge research, consulting, executive coaching, and advising to bear on the subject, Bapna and Ghose demystify the technology of AI itself. They offer a novel “House of AI” framework that encompasses traditional analytics, generative AI, and fair and ethical deployment of AI. Using examples from everyday life, they showcase how the modern AI-powered ecosystem fundamentally improves the emotional, physical, and material well-being of regular people across the globe. Thrive's mission is to educate the public about AI, shape realistic expectations, and foster informed discussions about a fast-emerging AI-shaped society.

In his bestselling first book, Eric Siegel explained how machine learning works. Now, in The AI Playbook, he shows how to capitalize on it.“Eric Siegel delivers a robust primer on machine learning, the key mechanism in AI. A forward-looking, practical book and a must-read for anyone in the information economy.”—Scott Galloway, NYU Stern Professor of Marketing; bestselling author of The Four“An antidote to today's relentless AI hype—why some AI initiatives thrive while others fail and what it takes for companies and people to succeed.”—Charles Duhigg, author of bestsellers The Power of Habit and Smarter Faster BetterThe greatest tool is the hardest to use. Machine learning is the world's most important general-purpose technology—but it's notoriously difficult to launch. Outside Big Tech and a handful of other leading companies, machine learning initiatives routinely fail to deploy, never realizing value. What's missing? A specialized business practice suitable for wide adoption. In The AI Playbook, bestselling author Eric Siegel presents the gold-standard, six-step practice for ushering machine learning projects from conception to deployment. He illustrates the practice with stories of success and of failure, including revealing case studies from UPS, FICO, and prominent dot-coms. This disciplined approach serves both sides: It empowers business professionals, and it establishes a sorely needed strategic framework for data professionals.Beyond detailing the practice, this book also upskills business professionals—painlessly. It delivers a vital yet friendly dose of semi-technical background knowledge that all stakeholders need to lead or participate in machine learning projects, end to end. This puts business and data professionals on the same page so that they can collaborate deeply, jointly establishing precisely what machine learning is called upon to predict, how well it predicts, and how its predictions are acted upon to improve operations. These essentials make or break each initiative—getting them right paves the way for machine learning's value-driven deployment.A note from the author: What kind of AI does this book cover? The buzzword AI can mean many things, but this book is about machine learning, which is a central basis for—and what many mean by—AI. To be specific, this book covers the most vital use cases of machine learning, those designed to improve a wide range of business operations.Endorsements:“Eric Siegel delivers a robust primer on machine learning, the key mechanism in AI. A forward-looking, practical book and a must-read for anyone in the information economy.” —Scott Galloway, NYU Stern Professor of Marketing; bestselling author of The Four“An antidote to today's relentless AI hype—why some AI initiatives thrive while others fail and what it takes for companies and people to succeed.”—Charles Duhigg, author of bestsellers The Power of Habit and Smarter Faster Better“Set aside the hype and focus on getting things to work in practice. This is a crisp, necessary, and deeply helpful guide to getting things done with AI. Essential reading.”—Mustafa Suleyman, Cofounder and CEO, Inflection AI; author of The Coming Wave“In this book, Eric brings machine learning to life and provides a roadmap for how to operationalize it in the real world.”—Will Lansing, CEO, FICO“This should be requisite reading for any professional serious about driving true value through the power of machine learning. Eric presents a pragmatic approach and it's not just about having the best algorithms—it's about ensuring you have a path to true productionalization at scale.”—Jon Francis, Chief Data and Analytics Officer, GM“The AI Playbook grabs you from the first p

The essential reader on the philosophical foundations and implications of artificial intelligence, now comprehensively updated for the twenty-first century.In the quarter century since the publication of John Haugeland's Mind Design II, computer scientists have hit many of their objectives for successful artificial intelligence. Computers beat chess grandmasters, driverless cars navigate streets, autonomous robots vacuum our homes, and ChatGPT answers existential queries in iambic pentameter on command. Engineering has made incredible strides. But have we made progress in understanding and building minds? Comprehensively updated by Carl Craver and Colin Klein to reflect the astonishing ubiquity of machine learning in modern life, Mind Design III offers an essential collection of classic and contemporary essays on the philosophical foundations and implications of artificial intelligence. Contributions from a diverse range of philosophers and computer scientists address the nature of computation, the nature of thought, and the question of whether computers can be made to think. With extensive new material reflecting the explosive growth and diversification of AI approaches, this classic reader equips students to assess the possibility of, and progress toward, building minds out of computers.New edition highlights: New chapters on advances in deep neural networks, reinforcement learning, and causal learningNew material on the complementary intersection of neuroscience and AIOrganized thematically rather than chronologically Brand new introductions to each section that include suggestions for coursework and further reading

How we can get more joy from our machines by telling them what our hearts desire.In this informative, accessible, and very funny book, Michael L. Littman inspires readers to learn how to tell machines what to do for us. Rather than give in to the fear that computers will steal our jobs, spy on us and control what we buy and whom we vote for, we can improve our relationship with them just by learning basic programming skills. Our devices will help us, Littman writes, if we can say what we want in a way they can understand. Each chapter of the book focuses on a particular element of what can be said, providing examples of how we use similar communication in our daily interactions with people. Littman offers ways readers can experiment with these ideas right away, using publicly available systems that might also make us more productive as a welcome side effect. Each chapter also reflects on how the use of these programming components can be expedited by machine learning. With humor and teacherly guidance, Code to Joy brings into view a future where programming is like reading—something everyone can learn.

A comprehensive introduction to the mathematical foundations of movement and actuation that apply equally to animals and machines.This textbook offers a computational framework for the sensorimotor stage of development as applied to robotics. Much work in developmental robotics is based on ad hoc examples, without a full computational basis. This book's comprehensive and complete treatment fills the gap, drawing on the principal mechanisms of development in the first year of life to introduce what is essentially an operating system for developing robots. The goal is to apply principles of development to robot systems that not only achieve new levels of performance but also provide evidence for scientific theories of human development.

How the use of machine learning to analyze art images has revived formalism in art history, presenting a golden opportunity for art historians and computer scientists to learn from one another.Though formalism is an essential tool for art historians, much recent art history has focused on the social and political aspects of art. But now art historians are adopting machine learning methods to develop new ways to analyze the purely visual in datasets of art images. Amanda Wasielewski uses the term “computational formalism” to describe this use of machine learning and computer vision technique in art historical research. At the same time that art historians are analyzing art images in new ways, computer scientists are using art images for experiments in machine learning and computer vision. Their research, says Wasielewski, would be greatly enriched by the inclusion of humanistic issues.The main purpose in applying computational techniques such as machine learning to art datasets is to automate the process of categorization using metrics such as style, a historically fraught concept in art history. After examining a fifteen-year trajectory in image categorization and art dataset creation in the fields of machine learning and computer vision, Wasielewski considers deep learning techniques that both create and detect forgeries and fakes in art. She investigates examples of art historical analysis in the fields of computer and information sciences, placing this research in the context of art historiography. She also raises questions as which artworks are chosen for digitization, and of those artworks that are born digital, which works gain acceptance into the canon of high art.

An introduction to how neuroethology can inform the development of robots controlled by synaptic networks instead of algorithms, from a pioneer in biorobotics.The trait most fundamental to the evolution of animals is the capability to adapt to novel circumstances in unpredictable environments. Recent advances in biomimetics have made it feasible to construct robots modeled on such unsupervised autonomous behavior, and animal models provide a library of existence proofs. Filling an important gap in the field, this introductory textbook illuminates how neurobiological principles can inform the development of robots that are controlled by synaptic networks, as opposed to algorithms. Joseph Ayers provides a comprehensive overview of the sensory and motor systems of a variety of model biological systems and shows how their behaviors may be implemented in artificial systems, such as biomimetic robots. Introduces the concept of biological intelligence as applied to robots, building a strategy for autonomy based on the neuroethology of simple animal modelsProvides a mechanistic physiological framework for the control of innate behaviorIllustrates how biomimetic vehicles can be operated in the field persistently and adaptivelyDeveloped by a pioneer in biorobotics with decades of teaching experienceProven in the classroom Suitable for professionals and researchers as well as undergraduate and graduate students in cognitive science and computer science

The power of the ever-increasing tools and algorithms for prediction and their paradoxical effects on risk.The Age of Prediction is about two powerful, and symbiotic, trends: the rapid development and use of artificial intelligence and big data to enhance prediction, as well as the often paradoxical effects of these better predictions on our understanding of risk and the ways we live. Beginning with dramatic advances in quantitative investing and precision medicine, this book explores how predictive technology is quietly reshaping our world in fundamental ways, from crime fighting and warfare to monitoring individual health and elections.As prediction grows more robust, it also alters the nature of the accompanying risk, setting up unintended and unexpected consequences. The Age of Prediction details how predictive certainties can bring about complacency or even an increase in risks—genomic analysis might lead to unhealthier lifestyles or a GPS might encourage less attentive driving. With greater predictability also comes a degree of mystery, and the authors ask how narrower risks might affect markets, insurance, or risk tolerance generally. Can we ever reduce risk to zero? Should we even try? This book lays an intriguing groundwork for answering these fundamental questions and maps out the latest tools and technologies that power these projections into the future, sometimes using novel, cross-disciplinary tools to map out cancer growth, people's medical risks, and stock dynamics.

The first comprehensive guide to distributional reinforcement learning, providing a new mathematical formalism for thinking about decisions from a probabilistic perspective.Distributional reinforcement learning is a new mathematical formalism for thinking about decisions. Going beyond the common approach to reinforcement learning and expected values, it focuses on the total reward or return obtained as a consequence of an agent's choices—specifically, how this return behaves from a probabilistic perspective. In this first comprehensive guide to distributional reinforcement learning, Marc G. Bellemare, Will Dabney, and Mark Rowland, who spearheaded development of the field, present its key concepts and review some of its many applications. They demonstrate its power to account for many complex, interesting phenomena that arise from interactions with one's environment.The authors present core ideas from classical reinforcement learning to contextualize distributional topics and include mathematical proofs pertaining to major results discussed in the text. They guide the reader through a series of algorithmic and mathematical developments that, in turn, characterize, compute, estimate, and make decisions on the basis of the random return. Practitioners in disciplines as diverse as finance (risk management), computational neuroscience, computational psychiatry, psychology, macroeconomics, and robotics are already using distributional reinforcement learning, paving the way for its expanding applications in mathematical finance, engineering, and the life sciences. More than a mathematical approach, distributional reinforcement learning represents a new perspective on how intelligent agents make predictions and decisions.

Why robots defy our existing moral and legal categories and how to revolutionize the way we think about them.Robots are a curious sort of thing. On the one hand, they are technological artifacts—and thus, things. On the other hand, they seem to have social presence, because they talk and interact with us, and simulate the capabilities commonly associated with personhood. In Person, Thing, Robot, David J. Gunkel sets out to answer the vexing question: What exactly is a robot? Rather than try to fit robots into the existing categories by way of arguing for either their reification or personification, however, Gunkel argues for a revolutionary reformulation of the entire system, developing a new moral and legal ontology for the twenty-first century and beyond.In this book, Gunkel investigates how and why efforts to use existing categories to classify robots fail, argues that “robot” designates an irreducible anomaly in the existing ontology, and formulates an alternative that restructures the ontological order in both moral philosophy and law. Person, Thing, Robot not only addresses the issues that are relevant to students, teachers, and researchers working in the fields of moral philosophy, philosophy of technology, science and technology studies (STS), and AI/robot law and policy but it also speaks to controversies that are important to AI researchers, robotics engineers, and computer scientists concerned with the social consequences of their work.

An advanced book for researchers and graduate students working in machine learning and statistics who want to learn about deep learning, Bayesian inference, generative models, and decision making under uncertainty.An advanced counterpart to Probabilistic Machine Learning: An Introduction, this high-level textbook provides researchers and graduate students detailed coverage of cutting-edge topics in machine learning, including deep generative modeling, graphical models, Bayesian inference, reinforcement learning, and causality. This volume puts deep learning into a larger statistical context and unifies approaches based on deep learning with ones based on probabilistic modeling and inference. With contributions from top scientists and domain experts from places such as Google, DeepMind, Amazon, Purdue University, NYU, and the University of Washington, this rigorous book is essential to understanding the vital issues in machine learning.Covers generation of high dimensional outputs, such as images, text, and graphs Discusses methods for discovering insights about data, based on latent variable models Considers training and testing under different distributionsExplores how to use probabilistic models and inference for causal inference and decision makingFeatures online Python code accompaniment

An authoritative, accessible, and up-to-date treatment of deep learning that strikes a pragmatic middle ground between theory and practice.Deep learning is a fast-moving field with sweeping relevance in today's increasingly digital world. Understanding Deep Learning provides an authoritative, accessible, and up-to-date treatment of the subject, covering all the key topics along with recent advances and cutting-edge concepts. Many deep learning texts are crowded with technical details that obscure fundamentals, but Simon Prince ruthlessly curates only the most important ideas to provide a high density of critical information in an intuitive and digestible form. From machine learning basics to advanced models, each concept is presented in lay terms and then detailed precisely in mathematical form and illustrated visually. The result is a lucid, self-contained textbook suitable for anyone with a basic background in applied mathematics.Up-to-date treatment of deep learning covers cutting-edge topics not found in existing texts, such as transformers and diffusion modelsShort, focused chapters progress in complexity, easing students into difficult concepts Pragmatic approach straddling theory and practice gives readers the level of detail required to implement naive versions of modelsStreamlined presentation separates critical ideas from background context and extraneous detailMinimal mathematical prerequisites, extensive illustrations, and practice problems make challenging material widely accessible Programming exercises offered in accompanying Python Notebooks

An introduction to the intellectual foundations and practical utility of the recent work on fairness and machine learning.Fairness and Machine Learning introduces advanced undergraduate and graduate students to the intellectual foundations of this recently emergent field, drawing on a diverse range of disciplinary perspectives to identify the opportunities and hazards of automated decision-making. It surveys the risks in many applications of machine learning and provides a review of an emerging set of proposed solutions, showing how even well-intentioned applications may give rise to objectionable results. It covers the statistical and causal measures used to evaluate the fairness of machine learning models as well as the procedural and substantive aspects of decision-making that are core to debates about fairness, including a review of legal and philosophical perspectives on discrimination. This incisive textbook prepares students of machine learning to do quantitative work on fairness while reflecting critically on its foundations and its practical utility.• Introduces the technical and normative foundations of fairness in automated decision-making• Covers the formal and computational methods for characterizing and addressing problems• Provides a critical assessment of their intellectual foundations and practical utility• Features rich pedagogy and extensive instructor resources

A surprising vision of how human intelligence will coevolve with digital technology and revolutionize how we think and behave.It is natural for us to fear artificial intelligence. But does Siri really want to kill us? Perhaps we are falling into the trap of projecting human traits onto the machines we might build. In Evolutionary Intelligence, Neuman offers a surprisingly positive vision in which computational intelligence compensates for the well-recognized limits of human judgment, improves decision making, and actually increases our agency. In artful, accessible, and adventurous prose, Neuman takes the reader on an exciting, fast-paced ride, all the while making a convincing case about a revolution in computationally augmented human intelligence.Neuman argues that, just as the wheel made us mobile and machines made us stronger, the migration of artificial intelligence from room-sized computers to laptops to our watches, smart glasses, and even smart contact lenses will transform day-to-day human decision making. If intelligence is the capacity to match means with ends, then augmented intelligence can offer the ability to adapt to changing environments as we face the ultimate challenge of long-term survival.Tapping into a global interest in technology's potential impacts on society, economics, and culture, Evolutionary Intelligence demonstrates that our future depends on our ability to computationally compensate for the limitations of a human cognitive system that has only recently graduated from hunting and gathering.

An insightful investigation into the mechanisms underlying the predictive functions of neural networks—and their ability to chart a new path for AI.Prediction is a cognitive advantage like few others, inherently linked to our ability to survive and thrive. Our brains are awash in signals that embody prediction. Can we extend this capability more explicitly into synthetic neural networks to improve the function of AI and enhance its place in our world? Gradient Expectations is a bold effort by Keith L. Downing to map the origins and anatomy of natural and artificial neural networks to explore how, when designed as predictive modules, their components might serve as the basis for the simulated evolution of advanced neural network systems.Downing delves into the known neural architecture of the mammalian brain to illuminate the structure of predictive networks and determine more precisely how the ability to predict might have evolved from more primitive neural circuits. He then surveys past and present computational neural models that leverage predictive mechanisms with biological plausibility, identifying elements, such as gradients, that natural and artificial networks share. Behind well-founded predictions lie gradients, Downing finds, but of a different scope than those that belong to today's deep learning. Digging into the connections between predictions and gradients, and their manifestation in the brain and neural networks, is one compelling example of how Downing enriches both our understanding of such relationships and their role in strengthening AI tools. Synthesizing critical research in neuroscience, cognitive science, and connectionism, Gradient Expectations offers unique depth and breadth of perspective on predictive neural-network models, including a grasp of predictive neural circuits that enables the integration of computational models of prediction with evolutionary algorithms.

An introduction to neurorobotics that presents approaches and design principles for developing intelligent autonomous systems grounded in biology and neuroscience.Neurorobotics is an interdisciplinary field that draws on artificial intelligence, cognitive sciences, computer science, engineering, psychology, neuroscience, and robotics. Because the brain is closely coupled to the body and situated in the environment, neurorobots—autonomous systems modeled after some aspect of the brain—offer a powerful tool for studying neural function and may also be a means for developing autonomous systems with intelligence that rivals that of biological organisms. This textbook introduces approaches and design principles for developing intelligent autonomous systems grounded in biology and neuroscience. It is written for anyone interested in learning about this topic and can be used in cognitive robotics courses for students in psychology, cognitive science, and computer science.Neurorobotics covers the background and foundations of the field, with information on early neurorobots, relevant principles of neuroscience, learning rules and mechanisms, and reinforcement learning and prediction; neurorobot design principles grounded in neuroscience and principles of neuroscience research; and examples of neurorobots for navigation, developmental robotics, and social robots, presented with the cognitive science and neuroscience background that inspired them. A supplementary website offers videos, robot simulations, and links to software repositories with neurorobot examples.

Stories from the future of intelligent machines—from rescue drones to robot spouses—and accounts of cutting-edge research that could make it all possible.Tech prognosticators promised us robots—autonomous humanoids that could carry out any number of tasks. Instead, we have robot vacuum cleaners. But, as Dario Floreano and Nicola Nosengo report, advances in robotics could bring those rosy predictions closer to reality. A new generation of robots, directly inspired by the intelligence and bodies of living organisms, will be able not only to process data but to interact physically with humans and the environment. In this book, Floreano, a roboticist, and Nosengo, a science writer, bring us tales from the future of intelligent machines—from rescue drones to robot spouses—along with accounts of the cutting-edge research that could make it all possible. These stories from the not-so-distant future show us robots that can be used for mitigating effects of climate change, providing healthcare, working with humans on the factory floor, and more. Floreano and Nosengo tell us how an application of swarm robotics could protect Venice from flooding, how drones could reduce traffic on the congested streets of mega-cities like Hong Kong, and how a “long-term relationship model” robot could supply sex, love, and companionship. After each fictional scenario, they explain the technologies that underlie it, describing advances in such areas as soft robotics, swarm robotics, aerial and mobile robotics, humanoid robots, wearable robots, and even biohybrid robots based on living cells. Robotics technology is no silver bullet for all the world's problems—but it can help us tackle some of the most pressing challenges we face.

Two management and technology experts show that AI is not a job destroyer, exploring worker-AI collaboration in real-world work settings.This book breaks through both the hype and the doom-and-gloom surrounding automation and the deployment of artificial intelligence-enabled—“smart”—systems at work. Management and technology experts Thomas Davenport and Steven Miller show that, contrary to widespread predictions, prescriptions, and denunciations, AI is not primarily a job destroyer. Rather, AI changes the way we work—by taking over some tasks but not entire jobs, freeing people to do other, more important and more challenging work. By offering detailed, real-world case studies of AI-augmented jobs in settings that range from finance to the factory floor, Davenport and Miller also show that AI in the workplace is not the stuff of futuristic speculation. It is happening now to many companies and workers.These cases include a digital system for life insurance underwriting that analyzes applications and third-party data in real time, allowing human underwriters to focus on more complex cases; an intelligent telemedicine platform with a chat-based interface; a machine learning-system that identifies impending train maintenance issues by analyzing diesel fuel samples; and Flippy, a robotic assistant for fast food preparation. For each one, Davenport and Miller describe in detail the work context for the system, interviewing job incumbents, managers, and technology vendors. Short “insight” chapters draw out common themes and consider the implications of human collaboration with smart systems.

An expert offers a guide to where we should use artificial intelligence—and where we should not.Before we know it, artificial intelligence (AI) will work its way into every corner of our lives, making decisions about, with, and for us. Is this a good thing? There's a tendency to think that machines can be more “objective” than humans—can make better decisions about job applicants, for example, or risk assessments. In Awkward Intelligence, AI expert Katharina Zweig offers readers the inside story, explaining how many levers computer and data scientists must pull for AI's supposedly objective decision making. She presents the good and the bad: AI is good at processing vast quantities of data that humans cannot—but it's bad at making judgments about people. AI is accurate at sifting through billions of websites to offer up the best results for our search queries and it has beaten reigning champions in games of chess and Go. But, drawing on her own research, Zweig shows how inaccurate AI is, for example, at predicting whether someone with a previous conviction will become a repeat offender. It's no better than simple guesswork, and yet it's used to determine people's futures. Zweig introduces readers to the basics of AI and presents a toolkit for designing AI systems. She explains algorithms, big data, and computer intelligence, and how they relate to one another. Finally, she explores the ethics of AI and how we can shape the process. With Awkward Intelligence. Zweig equips us to confront the biggest question concerning AI: where we should use it—and where we should not.

AI is revolutionizing the world. Here's how democracies can come out on top.Artificial intelligence is revolutionizing the modern world. It is ubiquitous—in our homes and offices, in the present and most certainly in the future. Today, we encounter AI as our distant ancestors once encountered fire. If we manage AI well, it will become a force for good, lighting the way to many transformative inventions. If we deploy it thoughtlessly, it will advance beyond our control. If we wield it for destruction, it will fan the flames of a new kind of war, one that holds democracy in the balance. As AI policy experts Ben Buchanan and Andrew Imbrie show in The New Fire, few choices are more urgent—or more fascinating—than how we harness this technology and for what purpose. The new fire has three sparks: data, algorithms, and computing power. These components fuel viral disinformation campaigns, new hacking tools, and military weapons that once seemed like science fiction. To autocrats, AI offers the prospect of centralized control at home and asymmetric advantages in combat. It is easy to assume that democracies, bound by ethical constraints and disjointed in their approach, will be unable to keep up. But such a dystopia is hardly preordained. Combining an incisive understanding of technology with shrewd geopolitical analysis, Buchanan and Imbrie show how AI can work for democracy. With the right approach, technology need not favor tyranny.

STAYING IN CHARGE: How do we navigate a world populated by algorithms that beat us in chess, find us romantic partners, and tell us to “turn right in 500 yards”? “Anyone worried about the age of AI will sleep better after reading this intelligent account” about the limits and dangers of technology (Publishers Weekly). Doomsday prophets of technology predict that robots will take over the world, leaving humans behind in the dust. Tech industry boosters think replacing people with software might make the world a better place—while tech industry critics warn darkly about surveillance capitalism. Despite their differing views of the future, they all seem to agree: machines will soon do everything better than humans. In How to Stay Smart in a Smart World, Gerd Gigerenzer shows why that's not true, and tells us how we can stay in charge in a world populated by algorithms. Machines powered by artificial intelligence are good at some things (playing chess), but not others (life-and-death decisions, or anything involving uncertainty). Gigerenzer explains why algorithms often fail at finding us romantic partners (love is not chess), why self-driving cars fall prey to the Russian Tank Fallacy, and how judges and police rely increasingly on nontransparent “black box” algorithms to predict whether a criminal defendant will reoffend or show up in court. He invokes Black Mirror, considers the privacy paradox (people want privacy but give their data away), and explains that social media get us hooked by programming intermittent reinforcement in the form of the “like” button. We shouldn't trust smart technology unconditionally, Gigerenzer tells us, but we shouldn't fear it unthinkingly, either.

A detailed and up-to-date introduction to machine learning, presented through the unifying lens of probabilistic modeling and Bayesian decision theory.This book offers a detailed and up-to-date introduction to machine learning (including deep learning) through the unifying lens of probabilistic modeling and Bayesian decision theory. The book covers mathematical background (including linear algebra and optimization), basic supervised learning (including linear and logistic regression and deep neural networks), as well as more advanced topics (including transfer learning and unsupervised learning). End-of-chapter exercises allow students to apply what they have learned, and an appendix covers notation. Probabilistic Machine Learning grew out of the author's 2012 book, Machine Learning: A Probabilistic Perspective. More than just a simple update, this is a completely new book that reflects the dramatic developments in the field since 2012, most notably deep learning. In addition, the new book is accompanied by online Python code, using libraries such as scikit-learn, JAX, PyTorch, and Tensorflow, which can be used to reproduce nearly all the figures; this code can be run inside a web browser using cloud-based notebooks, and provides a practical complement to the theoretical topics discussed in the book. This introductory text will be followed by a sequel that covers more advanced topics, taking the same probabilistic approach.

How we can create artificial intelligence with broad, robust common sense rather than narrow, specialized expertise.It's sometime in the not-so-distant future, and you send your fully autonomous self-driving car to the store to pick up your grocery order. The car is endowed with as much capability as an artificial intelligence agent can have, programmed to drive better than you do. But when the car encounters a traffic light stuck on red, it just sits there—indefinitely. Its obstacle-avoidance, lane-following, and route-calculation capacities are all irrelevant; it fails to act because it lacks the common sense of a human driver, who would quickly figure out what's happening and find a workaround. In Machines like Us, Ron Brachman and Hector Levesque—both leading experts in AI—consider what it would take to create machines with common sense rather than just the specialized expertise of today's AI systems. Using the stuck traffic light and other relatable examples, Brachman and Levesque offer an accessible account of how common sense might be built into a machine. They analyze common sense in humans, explain how AI over the years has focused mainly on expertise, and suggest ways to endow an AI system with both common sense and effective reasoning. Finally, they consider the critical issue of how we can trust an autonomous machine to make decisions, identifying two fundamental requirements for trustworthy autonomous AI systems: having reasons for doing what they do, and being able to accept advice. Both in the end are dependent on having common sense.

How a web-scale network of autonomous micromanagers can challenge the AI revolution and combat the high cost of quantitative business optimization.The artificial intelligence (AI) revolution is leaving behind small businesses and organizations that cannot afford in-house teams of data scientists. In Microprediction, Peter Cotton examines the repeated quantitative tasks that drive business optimization from the perspectives of economics, statistics, decision making under uncertainty, and privacy concerns. He asks what things currently described as AI are not “microprediction,” whether microprediction is an individual or collective activity, and how we can produce and distribute high-quality microprediction at low cost. The world is missing a public utility, he concludes, while companies are missing an important strategic approach that would enable them to benefit—and also give back. In an engaging, colloquial style, Cotton argues that market-inspired “superminds” are likely to be very effective compared with other orchestration mechanisms in the domain of microprediction. He presents an ambitious yet practical alternative to the expensive “artisan” data science that currently drains money from firms. Challenging the machine learning revolution and exposing a contradiction at its heart, he offers engineers a new liberty: no longer reliant on quantitative experts, they are free to create intelligent applications using general-purpose application programming interfaces (APIs) and libraries. He describes work underway to encourage this approach, one that he says might someday prove to be as valuable to businesses—and society at large—as the internet.

Fundamental theory and practical algorithms of weakly supervised classification, emphasizing an approach based on empirical risk minimization.Standard machine learning techniques require large amounts of labeled data to work well. When we apply machine learning to problems in the physical world, however, it is extremely difficult to collect such quantities of labeled data. In this book Masashi Sugiyama, Han Bao, Takashi Ishida, Nan Lu, Tomoya Sakai and Gang Niu present theory and algorithms for weakly supervised learning, a paradigm of machine learning from weakly labeled data. Emphasizing an approach based on empirical risk minimization and drawing on state-of-the-art research in weakly supervised learning, the book provides both the fundamentals of the field and the advanced mathematical theories underlying them. It can be used as a reference for practitioners and researchers and in the classroom.The book first mathematically formulates classification problems, defines common notations, and reviews various algorithms for supervised binary and multiclass classification. It then explores problems of binary weakly supervised classification, including positive-unlabeled (PU) classification, positive-negative-unlabeled (PNU) classification, and unlabeled-unlabeled (UU) classification. It then turns to multiclass classification, discussing complementary-label (CL) classification and partial-label (PL) classification. Finally, the book addresses more advanced issues, including a family of correction methods to improve the generalization performance of weakly supervised learning and the problem of class-prior estimation.

A proposal that we think about digital technologies such as machine learning not in terms of artificial intelligence but as artificial communication.Algorithms that work with deep learning and big data are getting so much better at doing so many things that it makes us uncomfortable. How can a device know what our favorite songs are, or what we should write in an email? Have machines become too smart? In Artificial Communication, Elena Esposito argues that drawing this sort of analogy between algorithms and human intelligence is misleading. If machines contribute to social intelligence, it will not be because they have learned how to think like us but because we have learned how to communicate with them. Esposito proposes that we think of “smart” machines not in terms of artificial intelligence but in terms of artificial communication. To do this, we need a concept of communication that can take into account the possibility that a communication partner may be not a human being but an algorithm—which is not random and is completely controlled, although not by the processes of the human mind. Esposito investigates this by examining the use of algorithms in different areas of social life. She explores the proliferation of lists (and lists of lists) online, explaining that the web works on the basis of lists to produce further lists; the use of visualization; digital profiling and algorithmic individualization, which personalize a mass medium with playlists and recommendations; and the implications of the “right to be forgotten.” Finally, she considers how photographs today seem to be used to escape the present rather than to preserve a memory.

An expert on mind considers how animals and smart machines measure up to human intelligence.Octopuses can open jars to get food, and chimpanzees can plan for the future. An IBM computer named Watson won on Jeopardy! and Alexa knows our favorite songs. But do animals and smart machines really have intelligence comparable to that of humans? In Bots and Beasts, Paul Thagard looks at how computers ('bots') and animals measure up to the minds of people, offering the first systematic comparison of intelligence across machines, animals, and humans. Thagard explains that human intelligence is more than IQ and encompasses such features as problem solving, decision making, and creativity. He uses a checklist of twenty characteristics of human intelligence to evaluate the smartest machines--including Watson, AlphaZero, virtual assistants, and self-driving cars--and the most intelligent animals--including octopuses, dogs, dolphins, bees, and chimpanzees. Neither a romantic enthusiast for nonhuman intelligence nor a skeptical killjoy, Thagard offers a clear assessment. He discusses hotly debated issues about animal intelligence concerning bacterial consciousness, fish pain, and dog jealousy. He evaluates the plausibility of achieving human-level artificial intelligence and considers ethical and policy issues. A full appreciation of human minds reveals that current bots and beasts fall far short of human capabilities.

How people judge humans and machines differently, in scenarios involving natural disasters, labor displacement, policing, privacy, algorithmic bias, and more.How would you feel about losing your job to a machine? How about a tsunami alert system that fails? Would you react differently to acts of discrimination depending on whether they were carried out by a machine or by a human? What about public surveillance? How Humans Judge Machines compares people's reactions to actions performed by humans and machines. Using data collected in dozens of experiments, this book reveals the biases that permeate human-machine interactions. Are there conditions in which we judge machines unfairly? Is our judgment of machines affected by the moral dimensions of a scenario? Is our judgment of machine correlated with demographic factors such as education or gender? César Hidalgo and colleagues use hard science to take on these pressing technological questions. Using randomized experiments, they create revealing counterfactuals and build statistical models to explain how people judge artificial intelligence and whether they do it fairly. Through original research, How Humans Judge Machines bring us one step closer tounderstanding the ethical consequences of AI.

A concise but informative overview of AI ethics and policy.Artificial intelligence, or AI for short, has generated a staggering amount of hype in the past several years. Is it the game-changer it's been cracked up to be? If so, how is it changing the game? How is it likely to affect us as customers, tenants, aspiring homeowners, students, educators, patients, clients, prison inmates, members of ethnic and sexual minorities, and voters in liberal democracies? Authored by experts in fields ranging from computer science and law to philosophy and cognitive science, this book offers a concise overview of moral, political, legal and economic implications of AI. It covers the basics of AI's latest permutation, machine learning, and considers issues such as transparency, bias, liability, privacy, and regulation.Both business and government have integrated algorithmic decision support systems into their daily operations, and the book explores the implications for our lives as citizens. For example, do we take it on faith that a machine knows best in approving a patient's health insurance claim or a defendant's request for bail? What is the potential for manipulation by targeted political ads? How can the processes behind these technically sophisticated tools ever be transparent? The book discusses such issues as statistical definitions of fairness, legal and moral responsibility, the role of humans in machine learning decision systems, “nudging” algorithms and anonymized data, the effect of automation on the workplace, and AI as both regulatory tool and target.

Experts from disciplines that range from computer science to philosophy consider the challenges of building AI systems that humans can trust.Artificial intelligence–based algorithms now marshal an astonishing range of our daily activities, from driving a car (“turn left in 400 yards”) to making a purchase (“products recommended for you”). How can we design AI technologies that humans can trust, especially in such areas of application as law enforcement and the recruitment and hiring process? In this volume, experts from a range of disciplines discuss the ethical and social implications of the proliferation of AI systems, considering bias, transparency, and other issues.The contributors, offering perspectives from computer science, engineering, law, and philosophy, first lay out the terms of the discussion, considering the “ethical debts” of AI systems, the evolution of the AI field, and the problems of trust and trustworthiness in the context of AI. They go on to discuss specific ethical issues and present case studies of such applications as medicine and robotics, inviting us to shift the focus from the perspective of a “human-centered AI” to that of an “AI-decentered humanity.” Finally, they consider the future of AI, arguing that, as we move toward a hybrid society of cohabiting humans and machines, AI technologies can become humanity's allies.ContributorsFrancesco Amigoni, Federico Cabitza, Nello Cristianini, Mireille Hildebrandt, Marcello Pelillo, Gernot Rieder, Teresa Scantamburlo, Viola Schiaffonati, Judith Simon, Katherine Strandburg, Jennifer Wortman Vaughan, Hanna Wallach, Robert Williamson, Pak-Hang Wong

MIT presents a concise primer on machine learning—computer programs that learn from data and the basis of applications like voice recognition and driverless cars. No in-depth knowledge of math or programming required! Today, machine learning underlies a range of applications we use every day, from product recommendations to voice recognition—as well as some we don't yet use every day, including driverless cars. It is the basis for a new approach to artificial intelligence that aims to program computers to use example data or past experience to solve a given problem. In this volume in the MIT Press Essential Knowledge series, Ethem Alpaydin offers a concise and accessible overview of “the new AI.” This expanded edition offers new material on such challenges facing machine learning as privacy, security, accountability, and bias. Alpaydin explains that as Big Data has grown, the theory of machine learning—the foundation of efforts to process that data into knowledge—has also advanced. He covers: • The evolution of machine learning• Important learning algorithms and example applications• Using machine learning algorithms for pattern recognition• Artificial neural networks inspired by the human brain• Algorithms that learn associations between instances• Reinforcement learning• Transparency, explainability, and fairness in machine learning• The ethical and legal implicates of data-based decision making A comprehensive introduction to machine learning, this book does not require any previous knowledge of mathematics or programming—making it accessible for everyday readers and easily adoptable for classroom syllabi.

A general formal theory of causal reasoning as a logical study of causal models, reasoning, and inference.In this book, Alexander Bochman presents a general formal theory of causal reasoning as a logical study of causal models, reasoning, and inference, basing it on a supposition that causal reasoning is not a competitor of logical reasoning but its complement for situations lacking logically sufficient data or knowledge. Bochman also explores the relationship of this theory with the popular structural equation approach to causality proposed by Judea Pearl and explores several applications ranging from artificial intelligence to legal theory, including abduction, counterfactuals, actual and proximate causality, dynamic causal models, and reasoning about action and change in artificial intelligence.As logical preparation, before introducing causal concepts, Bochman describes an alternative, situation-based semantics for classical logic that provides a better understanding of what can be captured by purely logical means. He then presents another prerequisite, outlining those parts of a general theory of nonmonotonic reasoning that are relevant to his own theory. These two components provide a logical background for the main, two-tier formalism of the causal calculus that serves as the formal basis of his theory. He presents the main causal formalism of the book as a natural generalization of classical logic that allows for causal reasoning. This provides a formal background for subsequent chapters. Finally, Bochman presents a generalization of causal reasoning to dynamic domains.

Recent advances in the area of lifted inference, which exploits the structure inherent in relational probabilistic models.Statistical relational AI (StaRAI) studies the integration of reasoning under uncertainty with reasoning about individuals and relations. The representations used are often called relational probabilistic models. Lifted inference is about how to exploit the structure inherent in relational probabilistic models, either in the way they are expressed or by extracting structure from observations. This book covers recent significant advances in the area of lifted inference, providing a unifying introduction to this very active field.After providing necessary background on probabilistic graphical models, relational probabilistic models, and learning inside these models, the book turns to lifted inference, first covering exact inference and then approximate inference. In addition, the book considers the theory of liftability and acting in relational domains, which allows the connection of learning and reasoning in relational domains.ContributorsBabak Ahmadi, Hendrik Blockeel, Hung Bui, Yuqiao Chen, Arthur Choi, Jaesik Choi, Adnan Darwiche, Jesse Davis, Rodrigo de Salvo Braz, Pedro Domingos, Daan Fierens, Martin Grohe, Fabian Hadiji, Seyed Mehran Kazemi, Kristian Kersting, Roni Khardon, Angelika Kimmig, Jacek Kisynski, Daniel Lowd, Wannes Meert, Martin Mladenov, Raymond Mooney, Sriraam Natarajan, Mathias Niepert, David Poole, Scott Sanner, Pascal Schweitzer, Nima Taghipour, Guy Van den Broeck

How science fiction has been a tool for understanding and living through rapid technological change.The world today seems to be slipping into a science fiction future. We have phones that speak to us, cars that drive themselves, and connected devices that communicate with each other in languages we don't understand. Depending on the news of the day, we inhabit either a technological utopia or a Brave New World nightmare. This volume in the MIT Press Essential Knowledge series surveys the uses of science fiction. It focuses on what is at the core of all definitions of science fiction: a vision of the world made otherwise and what possibilities might flow from such otherness.After a brief overview of the genre's origins, science fiction authority Sherryl Vint considers how and why contemporary science fiction is changing. She explores anxieties in current science fiction over such key sites of technological innovation as artificial intelligence, genomic research and commodified biomedicine, and climate change. Connecting science fiction with speculative design and futurology in the corporate world, she argues that science fiction does not merely reflect these trends, but has a role in directing them.