Brief announcement: coupling for Markov decision processes - application to self-stabilization with arbitrary schedulers.
Laurent FribourgStéphane MessikaPublished in: PODC (2005)
Keyphrases
- markov decision processes
- finite state
- reinforcement learning
- model based reinforcement learning
- reachability analysis
- optimal policy
- finite horizon
- state space
- reinforcement learning algorithms
- decision theoretic planning
- infinite horizon
- average reward
- markov decision process
- action sets
- policy iteration
- model checking
- decision processes
- planning under uncertainty
- probabilistic planning
- policy evaluation
- semi markov decision processes
- dynamic programming