Stochastic Bisimulation for MDPs Using Reachability Analysis.
Felipe Martins dos SantosLeliane Nunes de BarrosMijail Gamarra HolguinPublished in: BRACIS (2013)
Keyphrases
- reachability analysis
- markov decision processes
- model checking
- state space
- timed automata
- incremental algorithms
- reinforcement learning
- finite state
- policy iteration
- optimal policy
- reinforcement learning algorithms
- continuous state spaces
- finite horizon
- average reward
- planning under uncertainty
- partially observable
- markov decision process
- monte carlo
- dynamic programming