Introduction to Special Section on Evolutionary Computation in Games

Abstract

Ever since the dawn of artificial intelligence in the 1950s, games have been part and parcel of this lively field. In 1957, a year after the Dartmouth Conference that marked the official birth of AI, Alex Bernstein designed a program for the IBM 704 that played two amateur games of chess. In 1958, Allen Newell, J. C. Shaw, and Herbert Simon introduced a more sophisticated chess program (beaten in 35 moves by a 10-year-old beginner in its last official game played in 1960). Arthur L. Samuel of IBM spent much of the fifties working on game-playing AI programs, and by 1961 he had a checkers program that could play rather decently. In 1961 and 1963 Donald Michie described a simple trial-and-error learning system for learning how to play Tic-Tac-Toe (or Noughts and Crosses) called MENACE (for Matchbox Educable Noughts and Crosses Engine). These are but examples of highly popular games that have been treated by AI researchers since the field’s inception. ‘‘There are two principal reasons to continue to do research on games,’’ wrote Epstein [1]. ‘‘First, human fascination with game playing is long-standing and pervasive. Anthropologists have catalogued popular games in almost every culture... Games intrigue us because they address important cognitive functions... The second reason to continue game-playing research is that some difficult games remain to be won, games that people play very well but computers do not. These games clarify