HQ-Learning: Discovering Markovian Subgoals for non-Markovian Reinforcement Learning

DSpace/Manakin Repository

HQ-Learning: Discovering Markovian Subgoals for non-Markovian Reinforcement Learning
 
Wiering, M.A.; Schmidhuber, J.
(1996) Technical report IDSIA-95-96, pp. 1 - 13
(Report)
Download/Full Text
Open Access version via Utrecht University Repository
 
Keywords: reinforcement learning, hierarchical Q-learning, POMDPs, non-Markovian interfaces, subgoal learning
Publisher: IDSIA
 
 
See more statistics about this item