Limit Synchronization in Markov Decision Processes.
Laurent DoyenThierry MassartMahsa ShirmohammadiPublished in: CoRR (2013)
Keyphrases
- markov decision processes
- state space
- reinforcement learning
- finite state
- optimal policy
- policy iteration
- transition matrices
- dynamic programming
- partially observable
- infinite horizon
- average reward
- reachability analysis
- reward function
- planning under uncertainty
- risk sensitive
- average cost
- decision processes
- reinforcement learning algorithms
- decision theoretic planning
- factored mdps
- decision diagrams
- model based reinforcement learning
- action sets
- state and action spaces
- action space
- finite horizon
- markov decision process
- long run
- search space