Table of Contents
Series page
Title page
Copyright page
On Thinking Playfully
Prologue: AI&I
What Is This Book?
1
In the Beginning of AI, There Were Games
2
Do You Need to Be Intelligent to Play Games?
3
What Is (Artificial) Intelligence?
4
Do Video Games Have Artificial Intelligence?
5
Growing a Mind and Learning to Play
6
Do Games Learn from You When You Play Them?
7
Automating Creativity
8
Designing for AI
9
General Intelligence and Games in General
10
Synthesis
Further Reading
Bibliography
Index
List of tables
Table 2.1 The various cognitive abilities according to Cattell-Horn-Carroll theory and some examples of their use in games
List of figures
Figure 1.1 Chess existed for thousands of years before it became central to artificial intelligence research. (Courtesy of Wikimedia Commons under a Creative Commons 3.0 license.)
Figure 1.2 Go, the simpler but harder (for a computer) Asian cousin of Chess. (Photo by Linh Nguyen under Creative Commons 2.0 license.)
Figure 2.1 The genre-defining platform game Super Mario Bros. (Nintendo, 1985).
Figure 2.2 Angry Birds (Rovio, 2009), the physics puzzler that was on seemingly everyone’s iPhone after it debuted.
Figure 2.3 A planning algorithm (a version of the A* algorithm, discussed in chapter 4) playing a clone of Super Mario Bros. The black lines show the various future paths the algorithm is considering.
Figure 4.1 First person-shooters are so called because you view the world from a first-person perspective and, well, shoot things. Call of Duty: Modern Warfare 2 (Infinity Ward, 2009) is a good representative of the genre.
Figure 4.2 The games in the Civilization series (Firaxis, 1991–2016) allow you to lead a civilization through thousands of years of expansion, research, diplomacy, and war. The possibility space is quite overwhelming for computers and humans alike.
Figure 5.1 This figure illustrates a very simple neural network. It is organized into four layers: an input layer (with six nodes), two hidden layers (with four and three neurons each), and an output layer (with only one neuron). Each node (often called a “neuron” by analogy to biological neurons in brains) belongs to a particular layer and is connected to all neurons in the next layer. This type of neural network is called a feedforward network, because values are fed (or propagated) forward from one layer to the next.
Figure 6.1 Balancing on a ledge in Tomb Raider: Underworld (Crystal Dynamics, 2008).
Figure 6.2 The cubistic world of Minecraft (Mojang, 2011). The game largely revolves around mining cubes for material so that you can build things out of other cubes.
Figure 6.3 World of Warcraft (Blizzard, 2004) is a massively multiplayer online role-playing game; much of the game is communicating with other players over text or voice chat.
Figure 7.1 The racetracks that evolved from the neural networks.
Figure 7.2 Rogue (A. I. Design, 1980), the original roguelike, has modest hardware requirements because it was developed for a computer that had less computational power than your fridge. The smiley face represents the player character.
Figure 7.3 Approaching a space station in Elite (Acornsoft, 1984).
Figure 7.4 In No Man’s Sky (Hello Games, 2016) all planets are procedurally generated, including their flora, fauna, and geology.
Figure 7.5 Part of a level generated in the Mario AI framework using an evolutionary algorithm.
Figure 7.6 Diagram of a neural network that takes level design parameters and playing style as inputs, and outputs predicted player affect. By keeping playing style constant and optimizing for desired player affect, we can find out what types of levels would likely cause certain experiences in the player.
Figure 7.7 Yavalath (Nestorgames, 2007) was designed by the Ludi system, which was designed by Cameron Browne.
Figure 7.8 To That Sect (Michael Cook, 2014) was designed by the ANGELINA system, which was designed by Michael Cook.
Figure 7.9 Editing a map sketch with Sentient Sketchbook.
Figure 8.1 DOOM (id Software, 1993) was one of the original first-person shooters and a major influence in the development of this game genre.
Figure 8.2 In Third Eye Crime (Moonshot Games, 2014), the colors on the ground signal to the player both where the guards can currently see and where they are thinking of looking next, offering the player a view into the mind of the enemy.
Figure 8.3 A scene from Spy Party (Chris Hecker, 2009) features a number of NPCs in a bar, and one player must try to blend in seamlessly with them.
Figure 8.4 The giant creatures in Black and White (Lionhead Studios, 2001) can do your bidding, but only if you train them well.
Figure 8.5 A family tree of brains in EvoCommander (Daniel Jallov, 2015). Before a match, you choose which of your trained brains to bring with you into battle.
Figure 8.6 Evolved weapons in Galactic Arms Race (Evolutionary Games, 2009).
Figure 8.7 A romantic encounter in The Sims 4 (Maxis, 2013).
Figure 9.1 TORCS, 2014. (Image courtesy of the Libre Game Wiki.)
Figure 9.2 Four different games in the GVG-AI framework: Zelda, Butterflies, Boulder Dash, and Solar Fox. The common interface means that the same agent can play all games in the framework, but with varying skill.
Guide
Cover
Table of Contents
Begin Reading