ajedrezevolutivo

Upload: zorrin

Post on 14-Apr-2018

214 views

Category:

Documents


0 download

TRANSCRIPT

  • 7/27/2019 ajedrezevolutivo

    1/14

    Integrating CWU Chess in the ARENA Chess GUI

    Kyle Littlefield - CS Department

    Advisor: Razvan Andonie

    1

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    2/14

    Program Collaborators

    Ashur Odah (graduated 2004)

    Pushpinder Heer (graduated 2004)

    Joseph Lemnley (graduated 2006)

    Jonathan Widger (graduated 2006)

    Berk Erkul (graduated 2007)

    Lukas Magill (graduated 2008) Kyle Littlefield (currently enrolled)

    2

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    3/14

    The CWU Chess Program

    Genetic Algorithm trains 10 separate

    weights used in the neural network.

    Increases in skill level per set of 10games.

    3

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    4/14

    What is a Neural Network?

    It Starts with the basicunit called a Neuron

    Takes continuous inputfrom an external source

    Projects output based on

    the sum of the inputs andthreshold of the neuron.

    A neural network is excellent atclassifying an input to an output.

    Outputs the learned resultfrom the acquired input.

    4

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    5/14

    Criteria used by Neural Network

    Carrascal, A, Manrique, D, & Rios, J Neural networks evolutionary learning in chess game.

    5

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    6/14

    What is a Genetic

    Algorithm? Uses a population of

    chromosomes

    Each chromosomerepresents an individual.

    Has a fitness value representinghow well it survives in this world

    6

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    7/14

    The Goal

    To optimize the population

    Find the individuals with the highest fitness.

    Avoid getting caught in a local optima

    7

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    8/14

    The CWU Chess Program

    Genetic Algorithm trains 10 separate

    weights used in the neural network.

    Increases in skill level per set of 10games.

    Neural Network controls the behaviorof an Alpha-Beta Search algorithm.

    8

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    9/14

    What is an Alpha-Beta Search?

    A search algorithm

    Seeks to reduce the number of nodes

    evaluated in a search tree.

    8

    8

    8 5 2 9

    1

    0 1

    198

    Cut Off Region

    Node being

    Examined

    No need to evaluate the left branch of theexamined node.

    The value will be 1 or less.

    The root will be 8.

    9

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    10/14

    CWU Chess Program

    Running

    10

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    11/14

    The Problem

    The CWU-Chess program used a

    Windows Form for a User Interface.

    This didnt allow for communications with

    other chess interfaces or engines.

    The Universal Chess Interface (UCI)

    Protocol allowed this to happen.

    11

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    12/14

    The UCI Protocol

    Uses the console tocommunicatebetween the UserInterface and the

    Chess Engine.

    Communication isperformed by

    sending andreceiving textualinformation.

    12

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    13/14

    Implementation

    Remove the Windows Form Interface

    and all dependencies to it.

    Parse UCI Commands and parameters.

    Use them to control the engine.

    Redesign Genetic Algorithm fitnessfunction and network parameters.

    Redesign Genetic Algorithm fitnessfunction and network parameters.

    13

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University, Ellensburg, USA, May 21st 2009

  • 7/27/2019 ajedrezevolutivo

    14/14

    UCI Chess Running On

    ARENA

    14

    Symposium On University Research and CreativeExpression (SOURCE2009), Central Washington

    University Ellensburg USA May 21st 2009