Krzysztof Krawiec


Home

Research:

edit SideBar

In evolutionary learning of game-playing strategies, fitness evaluation is based on playing games with certain opponents. In this paper we investigate how the performance of these opponents and the way they are chosen influence the efficiency of learning. For this purpose we introduce a simple method for shaping the fitness function by sampling the opponents from a biased performance distribution. We compare the shaped function with existing fitness evaluation approaches that sample the opponents from an unbiased performance distribution or from a coevolving population. In an extensive computational experiment we employ these methods to learn Othello strategies and assess both the absolute and relative performance of the elaborated players. The results demonstrate the superiority of the shaping approach, and can be explained by means of performance profiles, an analytical tool that evaluate the evolved strategies using a range of variably skilled opponents.

@INPROCEEDINGS { Szubert:2013:SFF:2463372.2463513,
    AUTHOR = { Szubert, Marcin and Jaśkowski, Wojciech and Liskowski, Paweł and Krawiec, Krzysztof },
    TITLE = { Shaping fitness function for evolutionary learning of game strategies },
    BOOKTITLE = { Proceeding of the fifteenth annual conference on Genetic and evolutionary computation conference },
    SERIES = { GECCO '13 },
    YEAR = { 2013 },
    ISBN = { 978-1-4503-1963-8 },
    LOCATION = { Amsterdam, The Netherlands },
    PAGES = { 1149--1156 },
    NUMPAGES = { 8 },
    URL = { http://doi.acm.org/10.1145/2463372.2463513 },
    DOI = { 10.1145/2463372.2463513 },
    ACMID = { 2463513 },
    PUBLISHER = { ACM },
    ADDRESS = { New York, NY, USA },
    KEYWORDS = { coevolution, fitness evaluation, othello, shaping },
    ABSTRACT = { In evolutionary learning of game-playing strategies, fitness evaluation is based on playing games with certain opponents. In this paper we investigate how the performance of these opponents and the way they are chosen influence the efficiency of learning. For this purpose we introduce a simple method for shaping the fitness function by sampling the opponents from a biased performance distribution. We compare the shaped function with existing fitness evaluation approaches that sample the opponents from an unbiased performance distribution or from a coevolving population. In an extensive computational experiment we employ these methods to learn Othello strategies and assess both the absolute and relative performance of the elaborated players. The results demonstrate the superiority of the shaping approach, and can be explained by means of performance profiles, an analytical tool that evaluate the evolved strategies using a range of variably skilled opponents. },
}


Powered by PmWiki