Computational intelligence in games (Q5928035)

The book presents a sample of the most recent research on the application of computational intelligence techniques in the field of game theory and implementation. Game theory is a branch of operational research dealing with decision theory in a competitive situation. It involves mathematical calculations and heuristics that are used to optimise the efficient lines of play and may become very complex depending on the type and size of the game. It is at this point where the recent advances in computational intelligence are techniques are expected to contribute to, as they involve the use of computers to enable machines to simulate human performance and provide low cost solutions to difficult problems. The book consists of 6 chapters (not seven as stated in the preface!). In the first chapter, Chen, Fanelli, Castellano, and Jain provide a general introduction to computational intelligence paradigms including knowledge representation, probability-based approaches, fuzzy logic neural networks, genetic algorithms and rough sets. In chapter two, Chellapilla and Fogel present a new approach of evolving a neural network that play checkers without human expertise. In chapter three, Cazenave presents a comparative study using the game of Hex of two techniques, the retrograde analysis of patterns and metaprogramming. In chapter four, Schraudolph, Dayan, and Sejnowski demonstrate how neural networks may be trained to evaluate positions in the game of Go via temporal difference learning. In chapter five, Wiering, Salustowicz, and Schmidhuber present the use of reinforcement learning to evolve soccer team strategies. Finally, in chapter six, Ishibuchi and Nakashima describe how fuzzy rule-based systems can be applied to a market selection game with many players and several markets. The book provides a useful sample of computational intelligence techniques in games. As such, it is well addressed to researchers, practitioners and students who are interested to develop game techniques in computational intelligence environments. The first chapter provides a short but comprehensive introduction to the fundamentals of computational intelligence and, since the other chapters are selfcontained, it is expected even for the unfamiliar readers to follow easily the presented subjects.