Richard S. Sutton Hamid Reza Maei Doina Precup Shalabh Bhatnagar David Silver Csaba Szepesvári Eric Wiewiora Fast gradient-descent methods for temporal-difference learning with linear function approximation. 125 2009 ICML http://doi.acm.org/10.1145/1553374.1553501 conf/icml/2009 db/conf/icml/icml2009.html#SuttonMPBSSW09