Number of the records: 1  

Rule-based Analysis of Behaviour Learned by Evolutionary and Reinforcement Algorithms

  1. 1.
    0311244 - ÚI 2009 RIV DE eng C - Conference Paper (international conference)
    Slušný, Stanislav - Neruda, Roman - Vidnerová, Petra
    Rule-based Analysis of Behaviour Learned by Evolutionary and Reinforcement Algorithms.
    [Analýza chování naučených algoritmy evolučního a posilovaného učení pomocí pravidel.]
    Advanced Intelligent Computing Theories and Applications With Aspects of Artificial Intelligence. Berlin: Springer, 2008 - (Huang, D.; Wunsch, D.; Levine, D.; Jo, K.), s. 284-291. Lecture Notes in Artificial Intelligence, 5227. ISBN 978-3-540-85983-3.
    [ICIC 2008. International Conference on Intelligent Computing /4./. Shanghai (CN), 15.09.2008-18.09.2008]
    R&D Projects: GA AV ČR 1ET100300419
    Institutional research plan: CEZ:AV0Z10300504
    Keywords : radial basis function networks * reinforcement learning * evolutionary robotics
    Subject RIV: IN - Informatics, Computer Science
    DOI: https://doi.org/10.1007/978-3-540-85984-0_35

    We study behavioural patterns learned by a robotic agent by means of two different control and adaptive approaches - a radial basis function neural network trained by evolutionary algorithm, and a traditional reinforcement Q-learning algorithm. In both cases, a set of rules controlling the agent is derived from the learned controllers, and these sets are compared. It is shown that both procedures lead to reasonable and compact, albeit rather different, rule sets.

    Studujeme vzorce chování robotického agenta učícího se dvěma mechanismy – sítěmi typu RBF učenými evolučním algoritmem, a tradičním algoritmem posilovaného Q-učení. V obou případech odvodíme množinu pravidel řízení agenta a porovnáme je. Oba přístupy vedou k racionálním a poměrně kompaktním různým sadám pravidel.
    Permanent Link: http://hdl.handle.net/11104/0162913
     
Number of the records: 1  

  This site uses cookies to make them easier to browse. Learn more about how we use cookies.