View Document Preview and Link Document Date: 2013-11-28 14:07:54 Open Document File Size: 1,87 MB Share Result on Facebook
City London / / Company Monte Carlo / / / Facility Neuroscience Unit University College / Monte-Carlo Tree Search Arthur Guez aguez@gatsby.ucl.ac.uk Gatsby Computational Neuroscience Unit University College / Computer Science University College / / IndustryTerm bamdp solution / forward search algorithm / point-based value iteration algorithm / Online Search Online search methods / tree search / described search algorithms / sample-based online search algorithm / mdp planning algorithms / actor-critic algorithm / model-based reinforcement learning algorithms / online search literature / forward search / bfs3 algorithm / search horizon / shortcut solution / search tree / tractable exact algorithm / classical solutions / online methods / search space / search effort / pomcp algorithm / approximation algorithms / search algorithms / sample-based search methods / / Organization AI Access Foundation / University College London / / Person Carlo Tree / P. However / Peter Dayan dayan / David Silver / Monte-Carlo Planner / Mansour / Monte-Carlo Tree / / Position hidden model / rt / representative / Forward / / PublishedMedium Journal of Artificial Intelligence Research / / Technology Neuroscience / actor-critic algorithm / point-based value iteration algorithm / bfs3 algorithm / artificial intelligence / forward search algorithm / model-based Bayesian RL algorithms / search algorithms / existing approximation algorithms / planning algorithm / mdp planning algorithms / sample-based online search algorithm / pomcp algorithm / 1.3 described search algorithms / simulation / previous Bayesian model-based reinforcement learning algorithms / 2.2 Approximate Bayes-Adaptive Algorithms / Bayesian RL algorithms / / SocialTag 
Journal of Artificial Intelligence Research883 Submitted 06/13; publishedScalable and Efficient Bayes-Adaptive Reinforcement Learning Based on Monte-Carlo Tree Search