Artificial intelligence research has seen a lot pr
Artificial intelligence research has seen a lot progress over the years, moving from simply understanding images and speech to actually detecting emotions, driving cars, searching the web, and playing games. Because of these advancements, two AI experts decided to write an Artificial Intelligence and Games book, which will serve as the first comprehensive textbook on application and use of AI in, and for, games.Georgios Yannakakis and Julian Togelius, authors of the book, have both been teaching and researching game artificial intelligence at graduate and undergraduate levels.Yannakakis is currently an associate professor at the Institute of Digital Games at the University of Malta. His research interests include everything from AI, affective computing, neural networks, and procedural content generation. Togelius is an associate professor at the Department of Computer Science and Engineering at the Tandon School of Engineer, New York University. His research interests include artificial intelligence techniques for making computer games better, and he’s researching how games can make AI smarter. Both Yannakakis and Togelius felt that a textbook on game AI was necessary to future students and beneficial to the learning objectives of their programs. “[Yannakakis] and I have been working in this research field since at least 2005,” said Togelius. “We were active at the very beginning of the field’s formation, and did a number of influential research contributions in for example procedural content generation and player modeling.” Togelius added that the book is not only built on their own research, but it also features a lot of research published in the IEEE Conference on Computational Intelligence and Games (CIG), the AAAI Conference Artificial Intelligence in Interactive Digital Entertainment (AIIDE) and the IEEE Transactions on Computational Intelligence and AI in Games (TCIAIG). They are also looking at work that’s being done in the industry and presented at conferences like the Game Developers Conference.While there are a couple of books out there that delve into some of these topics around artificial intelligence, Togelius said that these books are older and tend to focus more on the needs of the game industry. “Our book is fully up to date with academic research as well,” said Togelius. “The use of video games as research environments in AI research in academia (and in big companies such as Google DeepMind and Facebook AI Research) has exploded in recent years, and our book reflects those developments.”Togelius also said that unlike other books, he and Yannakakis devote plenty of space to discussing the role of AI in content generation and player modeling, not just in game-playing.“There is an increasing use of video games to develop and test AI, and there’s also a long-standing use of AI methods in games,” said Togelius. “These fields use very different methods, and don’t always understand (or even know of) each other. While we come from the academic perspective, we are doing our best to try to bridge this divide.”Right now, the book’s first public draft is available for review. Any suggestions or feedback will be accepted by no later than June 20. The book will include three main chapters around playing games, generating content, and modeling players. They will also have an introductory chapter with overviews of the field and summaries of key algorithms, with a few chapters trying to “stake out the future of the research field,” said Togelius.