114
click to vote
ESANN
2006
15 years 4 months ago
2006
We focus on neuro-dynamic programming methods to learn state-action value functions and outline some of the inherent problems to be faced, when performing reinforcement learning in...