Page 141 - 49A Field Guide to Genetic Programming

12.9 Entertainment and Computer Games 127

is finding or generating high-quality heuristics for a problem, for a certain
class of instances of a problem, or even for a particular instance.
GP has been very successfully used as a hyperheuristic. For example, GP
has evolved competitive SAT solvers (Bader-El-Den and Poli, 2007a,b; Fuku-
naga, 2002; Kibria and Li, 2006), state-of-the-art or better than state-of-the-
art bin packing algorithms (Burke, Hyde, and Kendall, 2006; Burke, Hyde,
Kendall, and Woodward, 2007; Poli, Woodward, and Burke, 2007), particle
swarm optimisers (Poli, Di Chio, and Langdon, 2005; Poli, Langdon, and
Holland, 2005), evolutionary algorithms (Oltean, 2005), and travelling sales-
man problem solvers (Keller and Poli, 2007a,b,c; Oltean and Dumitrescu,
2004).

12.9 3/ 4 Entertainment and Computer Games

Today, a major usage of computers is interactive games (Priesterjahn,
Kramer, Weimer, and Goebels, 2006). There has been some work on in-
corporating artificial intelligence into mainstream commercial games. The
software owners are not keen on explaining exactly how much AI they use
or giving away sensitive information on how they use AI. Work on GP and
games includes (Azaria and Sipper, 2005a; Langdon and Poli, 2005; Vowk,
Wait, and Schmidt, 2004) as well as the human-competitive game players
mentioned in Section 12.3, page 120. Funes reports experiments which at-
tracted thousands of people via the Internet who were entertained by evolved
Tron players (Funes, Sklar, Juille, and Pollack, 1998b).
Since 2004, the annual IEEE CEC conference has included sessions on
evolutionary computation in games. After chairing the IEEE Symposium
on Computational Intelligence and Games 2005, at Essex University, Si-
mon Lucas founded the IEEE Computational Intelligence Society’s Techni-
cal Committee on Games. GP features heavily in the Games TC’s activi-
ties having being applied to Othello, poker, backgammon, draughts, chess,
Ms Pac-Man, robotic football and radio controlled model car racing.

12.10 The Arts

Computers have long been used to create purely aesthetic artifacts. Much
of today’s computer art tends to ape traditional drawing and painting, pro-
ducing static pictures on a computer monitor. However, the immediate
advantage of the computer screen — movement — can also be exploited. In
both cases evolutionary computation can and has been exploited. Indeed,
with evolution’s capacity for unlimited variation, evolutionary computation
offers the artist the scope to produce ever changing works. Some artists
have also worked with sound.