BIROn - Birkbeck Institutional Research Online

    A partial analysis of minimaxing game trees with random leaf values

    Levene, Mark and Fenner, Trevor (1995) A partial analysis of minimaxing game trees with random leaf values. International Computer Games Association Journal 18 (1), pp. 20-33. ISSN 1389-6911.

    Full text not available from this repository.

    Abstract

    Random minimaxing, introduced by Beal and Smith, is the process of using a random static evaluation function for scoring the leaf nodes of a full-width game tree and then computing the best move using the standard minimax procedure. Their experiments using random minimaxing in chess showed that the strength of play increases with the depth of the lookahead. We investigate random minim axing combinatorially in order to obtain a theoretical justification for Beal and Smith’s experiments. In particular, we show that, with respect to chess, random minim axing with the depth of lookahead equal to two is ‘stronger’ than the same with its depth equal to unity, under the assumption that a move by the first player is better the more it restricts the second player’s choice of moves (i.e., his mobility). We conjecture that these results can be generalized for depths of lookahead greater than two.

    Metadata

    Item Type: Article
    School: School of Business, Economics & Informatics > Computer Science and Information Systems
    Depositing User: Sarah Hall
    Date Deposited: 22 Mar 2021 20:02
    Last Modified: 22 Mar 2021 20:02
    URI: https://eprints.bbk.ac.uk/id/eprint/43620

    Statistics

    Downloads
    Activity Overview
    0Downloads
    33Hits

    Additional statistics are available via IRStats2.

    Archive Staff Only (login required)

    Edit/View Item Edit/View Item