Procedural Content Generation via Machine Learning

This book surveys current and future approaches to generating video game content with machine learning or Procedural Content Generation via Machine Learning (PCGML). Machine learning is having a major impact on many industries, including the video game industry. PCGML addresses the use of computers to generate new types of content for video games (game levels, quests, characters, etc.) by learning from existing content. The authors illustrate how PCGML is poised to transform the video games industry and provide the first ever beginner-focused guide to PCGML. This book features an accessible introduction to machine learning topics, and readers will gain a broad understanding of currently employed PCGML approaches in academia and industry. The authors provide guidance on how best to set up a PCGML project and identify open problems appropriate for a research project or thesis. This book is written with machine learning and games novices in mind and includes discussions of practical and ethical considerations along with resources and guidance for starting a new PCGML project.

Addresses the growing academic interest in PCGML Demonstrates common pitfalls in PCGML projects and how to avoid them Provides resources and guidance for starting new PCGML projects

Autorentext
Matthew Guzdial, Ph.D, is an Assistant Professor in the Computing Science Department at the University of Alberta and a Canada CIFAR AI Chair at the Alberta Machine Intelligence Institute (Amii). His research focuses on the intersection of machine learning, creativity, and human-centered computing. He is a recipient of an Early Career Researcher Award from NSERC, a Unity Graduate Fellowship, and two best conference paper awards from the International Conference on Computational Creativity. His work has been featured in the BBC, WIRED, Popular Science, and Time.
Sam Snodgrass is an AI researcher at modl.ai, a game AI company focused on bringing state of the art game AI research from academia to the games industry. His research focuses on making PCGML more accessible to non-ML experts. This work includes making PCGML systems more adaptable and self-reliant, reducing the authorial burden of creating training data through domain blending, and building tools that allow for easier interactions with the underlying PCGML systems and their outputs. Through his work at modl.ai he has deployed several mixed-initiative PCGML tools into game studios to assist with level design and creation.

Adam Summerville is the lead AI engineer for Procedural Content Generation at The Molasses Flood, a CD Projekt studio. Prior to this, he was an assistant professor at California State Polytechnic University, Pomona. His research focuses on the intersection of artificial intelligence in games with a high-level goal of enabling experiences that would not be possible without artificial intelligence. This research ranges from procedural generation of levels, social simulation for games, and the use of natural language processing for gameplay. His work has been shown at the SF MoMA and SlamDance and won the audience choice award at IndieCade.

Klappentext

Introduction.- Classical PCG.- An Introduction of ML Through PCG.- PCGML Process Overview.- Constraint-based PCGML Approaches.- Probabilistic PCGML Approaches.- Neural Networks: Introduction.- Sequence-based DNN PCGML.- Grid-based DNN PCGML.- Reinforcement Learning PCG.- Mixed-Initiative PCGML.- Open Problems.- Resource and Conclusions.

Inhalt
Introduction.- Classical PCG.- An Introduction of ML Through PCG.- PCGML Process Overview.- Constraint-based PCGML Approaches.- Probabilistic PCGML Approaches.- Neural Networks: Introduction.- Sequence-based DNN PCGML.- Grid-based DNN PCGML.- Reinforcement Learning PCG.- Mixed-Initiative PCGML.- Open Problems.- Resource and Conclusions.