ECML
2003
Springer
14 years 20 days ago
2003
Springer
We present in this paper a method to introduce a priori knowledge into reinforcement learning using temporally extended actions. The aim of our work is to reduce the learning time ...