Q-Learning with Recurrent Neural Networks as a Controller for the Inverted Pendulum Problem.

Ahmet Onat Hajime Kita Yoshikazu Nishikawa Q-Learning with Recurrent Neural Networks as a Controller for the Inverted Pendulum Problem. 837-840 1998 conf/iconip/1998 ICONIP db/conf/iconip/iconip1998.html#OnatKN98