Abstract: This chapter focuses on game-playing and game-learning automata. A game is a sequence of choices, each choice being made from a number of discrete alternatives. Every sequence terminates in an outcome, and every outcome has a value. It is conventional to assign outcome values from the point of view of the opening player, so that in Chess, for example, there are three outcome values, which we may denote + 1 , 0 , and -1. A game of that type can be represented by a directed graph. Terminal nodes are labeled with the corresponding outcome values, on the presumption that a description of the position itself is all one needs to know to assign a value to it. The value of a position, even a terminal position, could be dependent on the route followed in arriving at it. If there was a rule in Chess that making more than 100 moves automatically loses the game, it would be fruitless for a player to point out, after making his 100th move, that he could force mate in one. To handle this complication, it is customary to depict a game not by a generalized graph, in which the nodes stand for positions, but by the special kind of graph known as a tree.
Publication Year: 1966
Publication Date: 1966-01-01
Language: en
Type: book-chapter
Indexed In: ['crossref']
Access and Citation
Cited By Count: 59
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot