Recurrent Policy Gradients Daan Wierstra a,∗ , Alexander F¨orster a , Jan Peters b J¨ urgen Schmidhuber a,c,d , a IDSIA, b Max - J. B. Long - Document - PDFSEARCH.IO

First Page	Meta Content
	Recurrent Policy Gradients Daan Wierstra a,∗ , Alexander F¨orster a , Jan Peters b J¨ urgen Schmidhuber a,c,d , a IDSIA, b Max Add to Reading List Document Date: 2009-05-25 10:34:02 Open Document File Size: 1,46 MB Share Result on Facebook Company LSTM Recurrent Neural Networks / Recurrent Neural Networks / / Country Germany / / / Facility Switzerland Planck Institute / University of Lugano / / IndustryTerm history-dependent baseline network / conventional neural networks / reinforcement learning algorithm / neural network / extended baseline network / car driving simulation / neural networks / policy gradient algorithms / large product / baseline network / / Organization Institut f¨ / Switzerland Planck Institute for Biological Cybernetics / Faculty of Informatics / University of Lugano / / Person Jan Peters / / Position return Rt / actor / rt / ot and reward rt / controller / Corresponding author / / Product RPG / / Technology policy gradient algorithms / neural network / SRV algorithm / machine learning / simulation / RL algorithm / reinforcement learning algorithm / PG algorithms / Recurrent Policy Gradient algorithm / / SocialTag Cybernetics Neuroscience Long short term memory Recurrent neural network Reinforcement learning Backpropagation Neural networks Machine learning Computational neuroscience

Recurrent Policy Gradients Daan Wierstra a,∗ , Alexander F¨orster a , Jan Peters b J¨ urgen Schmidhuber a,c,d , a IDSIA, b MaxAdd to Reading List