Hierarchical Solution of Markov Decision Processes using Macro-actions
Milos HauskrechtNicolas MeuleauLeslie Pack KaelblingThomas L. DeanCraig BoutilierPublished in: CoRR (2013)
Keyphrases
- markov decision processes
- macro actions
- state space
- reinforcement learning
- finite state
- optimal policy
- transition matrices
- decision theoretic planning
- dynamic programming
- planning under uncertainty
- policy iteration
- decision processes
- reinforcement learning algorithms
- action space
- partially observable
- average cost
- markov decision process
- average reward
- infinite horizon
- learning algorithm
- neural network
- reward function
- temporally extended
- machine learning