Truncating Temporal Differences: On the Efficient Implementation of TD(lambda) for Reinforcement Learning.

Pawel Cichosz Truncating Temporal Differences: On the Efficient Implementation of TD(lambda) for Reinforcement Learning. 287-318 1995 2 J. Artif. Intell. Res. (JAIR) db/journals/jair/jair2.html#Cichosz95