Machines That Learn To Play Games

The mind-set that has dominated the history of computer game playing relies on straightforward exploitation of the available computing power. The fact that a machine can explore millions of variations sooner than the sluggish human can wink an eye has inspired hopes that the mystery of intelligence can be cracked, or at least side-stepped, by sheer force. Decades of the steadily growing strength of computer programs have attested to the soundness of this approach. It is clear that deeper understanding can cut the amount of necessary calculations by orders of magnitude. The papers collected in this volume describe how to instill learning skills in game playing machines. The reader is asked to keep in mind that this is not just about games -- the possibility that the discussed techniques will be used in control systems and in decision support always looms in the background.

Summary Deep Learning and the Game of Go teaches you how to apply the power of deep learning to complex reasoning tasks by building a Go-playing AI. After exposing you to the foundations of machine and deep learning, you'll use Python to build a bot and then teach it the rules of the game. Foreword by Thore Graepel, DeepMind Purchase of the print book includes a free eBook in PDF, Kindle, and ePub formats from Manning Publications. About the Technology The ancient strategy game of Go is an incredible case study for AI. In 2016, a deep learning-based system shocked the Go world by defeating a world champion. Shortly after that, the upgraded AlphaGo Zero crushed the original bot by using deep reinforcement learning to master the game. Now, you can learn those same deep learning techniques by building your own Go bot! About the Book Deep Learning and the Game of Go introduces deep learning by teaching you to build a Go-winning bot. As you progress, you'll apply increasingly complex training techniques and strategies using the Python deep learning library Keras. You'll enjoy watching your bot master the game of Go, and along the way, you'll discover how to apply your new deep learning skills to a wide range of other scenarios! What's inside Build and teach a self-improving game AI Enhance classical game AI systems with deep learning Implement neural networks for deep learning About the Reader All you need are basic Python skills and high school-level math. No deep learning experience required. About the Author Max Pumperla and Kevin Ferguson are experienced deep learning specialists skilled in distributed systems and data science. Together, Max and Kevin built the open source bot BetaGo. Table of Contents PART 1 - FOUNDATIONS Toward deep learning: a machine-learning introduction Go as a machine-learning problem Implementing your first Go bot PART 2 - MACHINE LEARNING AND GAME AI Playing games with tree search Getting started with neural networks Designing a neural network for Go data Learning from data: a deep-learning bot Deploying bots in the wild Learning by practice: reinforcement learning Reinforcement learning with policy gradients Reinforcement learning with value methods Reinforcement learning with actor-critic methods PART 3 - GREATER THAN THE SUM OF ITS PARTS AlphaGo: Bringing it all together AlphaGo Zero: Integrating tree search with reinforcement learning

A hands-on, application-based introduction to machine learning and artificial intelligence (AI) that guides young readers through creating compelling AI-powered games and applications using the Scratch programming language. Machine learning (also known as ML) is one of the building blocks of AI, or artificial intelligence. AI is based on the idea that computers can learn on their own, with your help. Machine Learning for Kids will introduce you to machine learning, painlessly. With this book and its free, Scratch-based, award-winning companion website, you'll see how easy it is to add machine learning to your own projects. You don't even need to know how to code! As you work through the book you'll discover how machine learning systems can be taught to recognize text, images, numbers, and sounds, and how to train your models to improve their accuracy. You'll turn your models into fun computer games and apps, and see what happens when they get confused by bad data. You'll build 13 projects step-by-step from the ground up, including: • Rock, Paper, Scissors game that recognizes your hand shapes • An app that recommends movies based on other movies that you like • A computer character that reacts to insults and compliments • An interactive virtual assistant (like Siri or Alexa) that obeys commands • An AI version of Pac-Man, with a smart character that knows how to avoid ghosts NOTE: This book includes a Scratch tutorial for beginners, and step-by-step instructions for every project. Ages 12+

A new vision of the future of games and game design, enabled by AI. Can games measure intelligence? How will artificial intelligence inform games of the future? In Playing Smart, Julian Togelius explores the connections between games and intelligence to offer a new vision of future games and game design. Video games already depend on AI. We use games to test AI algorithms, challenge our thinking, and better understand both natural and artificial intelligence. In the future, Togelius argues, game designers will be able to create smarter games that make us smarter in turn, applying advanced AI to help design games. In this book, he tells us how. Games are the past, present, and future of artificial intelligence. In 1948, Alan Turing, one of the founding fathers of computer science and artificial intelligence, handwrote a program for chess. Today we have IBM's Deep Blue and DeepMind's AlphaGo, and huge efforts go into developing AI that can play such arcade games as Pac-Man. Programmers continue to use games to test and develop AI, creating new benchmarks for AI while also challenging human assumptions and cognitive abilities. Game design is at heart a cognitive science, Togelius reminds us—when we play or design a game, we plan, think spatially, make predictions, move, and assess ourselves and our performance. By studying how we play and design games, Togelius writes, we can better understand how humans and machines think. AI can do more for game design than providing a skillful opponent. We can harness it to build game-playing and game-designing AI agents, enabling a new generation of AI-augmented games. With AI, we can explore new frontiers in learning and play.

In this textbook the author takes as inspiration recent breakthroughs in game playing to explain how and why deep reinforcement learning works. In particular he shows why two-person games of tactics and strategy fascinate scientists, programmers, and game enthusiasts and unite them in a common goal: to create artificial intelligence (AI). After an introduction to the core concepts, environment, and communities of intelligence and games, the book is organized into chapters on reinforcement learning, heuristic planning, adaptive sampling, function approximation, and self-play. The author takes a hands-on approach throughout, with Python code examples and exercises that help the reader understand how AI learns to play. He also supports the main text with detailed pointers to online machine learning frameworks, technical details for AlphaGo, notes on how to play and program Go and chess, and a comprehensive bibliography. The content is class-tested and suitable for advanced undergraduate and graduate courses on artificial intelligence and games. It's also appropriate for self-study by professionals engaged with applications of machine learning and with games development. Finally it's valuable for any reader engaged with the philosophical implications of artificial and general intelligence, games represent a modern Turing test of the power and limitations of AI.

Written for the novice AI programmer, this text introduces the reader to techniques such as finite state machines, fuzzy logic, neural networks and many others in an easy-to-understand language, supported with code samples throughout the text.

What Is Artificial Intelligence Video Games Artificial intelligence (AI) is used in video games to develop responsive, adaptive, or intelligent behaviors, primarily in non-player characters (NPCs), that are akin to the intellect of humans. Since the beginning of the video game industry in the 1950s, artificial intelligence has been an essential component of the medium. Artificial intelligence (AI) in video games is a discrete topic that is distinct from AI in academic settings. Rather than serving the purposes of machine learning or decision making, it is designed to enhance the experience of game players. The concept of artificial intelligence (AI) opponents became very popular during the golden age of arcade video games. This concept manifested itself in the form of graduated difficulty levels, distinct movement patterns, and in-game events that were reliant on the player's input. The behavior of non-player characters (NPCs) in modern games is frequently governed by tried-and-true methods such as pathfinding and decision trees. Data mining and procedural content production are two examples of AI applications that are frequently utilized in methods that are not immediately obvious to the user. How You Will Benefit (I) Insights, and validations about the following topics: Chapter 1: Artificial intelligence in video games Chapter 2: Artificial intelligence Chapter 3: List of artificial intelligence projects Chapter 4: Video game programmer Chapter 5: Interactive storytelling Chapter 6: Outline of video games Chapter 7: Outline of artificial intelligence Chapter 8: General game playing Chapter 9: Dynamic game difficulty balancing Chapter 10: Machine learning in video games (II) Answering the public top questions about artificial intelligence video games. (III) Real world examples for the usage of artificial intelligence video games in many fields. (IV) 17 appendices to explain, briefly, 266 emerging technologies in each industry to have 360-degree full understanding of artificial intelligence video games' technologies. Who This Book Is For Professionals, undergraduate and graduate students, enthusiasts, hobbyists, and those who want to go beyond basic knowledge or information for any kind of artificial intelligence video games.

Reinforcement learning encompasses both a science of adaptive behavior of rational beings in uncertain environments and a computational methodology for finding optimal behaviors for challenging problems in control, optimization and adaptive behavior of intelligent agents. As a field, reinforcement learning has progressed tremendously in the past decade. The main goal of this book is to present an up-to-date series of survey articles on the main contemporary sub-fields of reinforcement learning. This includes surveys on partially observable environments, hierarchical task decompositions, relational knowledge representation and predictive state representations. Furthermore, topics such as transfer, evolutionary methods and continuous spaces in reinforcement learning are surveyed. In addition, several chapters review reinforcement learning methods in robotics, in games, and in computational neuroscience. In total seventeen different subfields are presented by mostly young experts in those areas, and together they truly represent a state-of-the-art of current reinforcement learning research. Marco Wiering works at the artificial intelligence department of the University of Groningen in the Netherlands. He has published extensively on various reinforcement learning topics. Martijn van Otterlo works in the cognitive artificial intelligence group at the Radboud University Nijmegen in The Netherlands. He has mainly focused on expressive knowledge representation in reinforcement learning settings.

This book constitutes the refereed proceedings of the 14th European Conference on Machine Learning, ECML 2003, held in Cavtat-Dubrovnik, Croatia in September 2003 in conjunction with PKDD 2003. The 40 revised full papers presented together with 4 invited contributions were carefully reviewed and, together with another 40 ones for PKDD 2003, selected from a total of 332 submissions. The papers address all current issues in machine learning including support vector machine, inductive inference, feature selection algorithms, reinforcement learning, preference learning, probabilistic grammatical inference, decision tree learning, clustering, classification, agent learning, Markov networks, boosting, statistical parsing, Bayesian learning, supervised learning, and multi-instance learning.

This book constitutes the refereed preceedings of the 13th European Conference on Machine Learning, ECML 2002, held in Helsinki, Finland in August 2002. The 41 revised full papers presented together with 4 invited contributions were carefully reviewed and selected from numerous submissions. Among the topics covered are computational discovery, search strategies, Classification, support vector machines, kernel methods, rule induction, linear learning, decision tree learning, boosting, collaborative learning, statistical learning, clustering, instance-based learning, reinforcement learning, multiagent learning, multirelational learning, Markov decision processes, active learning, etc.