Daishi Harada Reinforcement Learning with Time. 577-582 1997 AAAI/IAAI db/conf/aaai/aaai97.html#Harada97