Fitted natural actor-critic: A new algorithm for continuous state-action MDPs. Melo, F. S. & Lopes, M. In Proc.\ European Conf.\ Machine Learning and Principles and Practive of Knowledge Discovery in Databases, pages 66-81, 2008.
Fitted natural actor-critic: A new algorithm for continuous state-action MDPs [pdf]Paper  bibtex   

Downloads: 0