| Document Date: 2009-01-13 06:30:39 Open Document File Size: 3,16 MBShare Result on Facebook
City Champaign / Tübingen / / Company Prentice-Hall / Neural Information Processing Systems / Ball / Artificial Neural Networks / MIT Press / John Wiley & Sons / / Country Germany / Jordan / / / Facility Jan Peters Max Planck Institute / / IndustryTerm statistical gradient-following algorithms / presented reinforcement learning algorithm / analytic solutions / Dynamics systems / online learning / arbitrary solution / Policy search / cross-products / good final solution / Real robot applications / policy search method / policy search methods / policy learning algorithm / probabilistic networks / mountain-car problem / dynamical systems / learning algorithms / / Organization MIT / Institute for Biological Cybernetics Spemannstr / / Person R. J. Williams / Finite Difference Gradients / / Position teacher / human player / actor / Fisher information matrix / rt / Episodic Natural Actor Critic / first author / Natural Actor Critic / reward rt / / Product Bang & Olufsen Form 2 Headphone/Headset / / ProvinceOrState New Jersey / Illinois / / PublishedMedium Machine Learning / / SportsEvent Ball-in-a-Cup / the Ball-in-a-Cup / the children’s game Ball-in-a-Cup / / Technology resulting algorithm / 1 3 Algorithm / artificial intelligence / PoWER algorithm / dom / machine learning / simulation / statistical gradient-following algorithms / Expectation-Maximization algorithms / EM Algorithm / EM-inspired algorithm / presented reinforcement learning algorithm / policy learning algorithm / / URL http /
SocialTag |