• Login
    View Item 
    •   DSpace Home
    • BIBLIOTECA DIGITAL: Teses & Dissertações
    • 40001016034P5 Programa de Pós-Graduação em Informática
    • Dissertações
    • View Item
    •   DSpace Home
    • BIBLIOTECA DIGITAL: Teses & Dissertações
    • 40001016034P5 Programa de Pós-Graduação em Informática
    • Dissertações
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Reinforcement learning for macro-management in the microrts game

    Thumbnail
    View/Open
    R - D - RAFAEL GASPARIN URBANO.pdf (1.314Mb)
    Date
    2019
    Author
    Urbano, Rafael Gasparin, 1992-
    Metadata
    Show full item record
    Subject
    Aprendizado do computador
    Inteligência artificial
    Ciência da Computação
    xmlui.dri2xhtml.METS-1.0.item-type
    Dissertação Digital
    Abstract
    Resumo: Nesta dissertacao, apresentamos uma proposta para a aplicacao de tecnicas de aprendizado por reforco em jogos de estrategia em tempo real para aprender estrategias de macro-gerenciamento. Esta proposta tem como objetivo colocar tecnicas comumente usadas em outros tipos de aplicacoes e mostrar como elas sao mais rapidas e eficientes do que outras tecnicas usadas neste campo. Para a implementacao, usamos um jogo chamado MicroRTS, que tem as caracteristicas minimas para ser considerado um jogo de Estrategia em Tempo Real e foi desenvolvido para avaliar implementacoes de inteligencia artificial sem precisar lidar com um jogo completo como StarCraft ou Wargus imediatamente. Palavras-chave: Aprendizado de Maquina, Aprendizado por reforco, MicroRTS.
     
    Abstract: In this dissertation, we present a proposal for applying reinforcement learning techniques in Real Time Strategy games to learn macro-management strategies. This proposal aims to put techniques that are commonly used in other types of applications and show how it is faster and more efficient than other techniques used in this field. For the implementation, we use a game called MicroRTS, which has the minimal characteristics to be considered an RTS game and was developed to evaluate Artificial Intelligence implementations without having to deal with a full game like StarCraft or Wargus immediately. Keywords: Machine Learning, Reinforcement Learning, MicroRTS, Macro-management
     
    URI
    https://hdl.handle.net/1884/65965
    Collections
    • Dissertações [348]

    DSpace software copyright © 2002-2016  DuraSpace
    Contact Us | Send Feedback
    Theme by 
    Atmire NV
     

     

    Browse

    All of DSpaceCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsxmlui.ArtifactBrowser.Navigation.browse_typeThis CollectionBy Issue DateAuthorsTitlesSubjectsxmlui.ArtifactBrowser.Navigation.browse_type

    My Account

    LoginRegister

    Statistics

    View Usage Statistics

    DSpace software copyright © 2002-2016  DuraSpace
    Contact Us | Send Feedback
    Theme by 
    Atmire NV