On the Markovian Randomized Strategy of Controller for Markov Decision Processes.
Taolue ChenTingting HanJian LuPublished in: FSKD (2006)
Keyphrases
- markov decision processes
- optimal policy
- policy iteration algorithm
- finite state
- policy iteration
- state space
- reinforcement learning
- transition matrices
- dynamic programming
- planning under uncertainty
- factored mdps
- average cost
- decision theoretic planning
- decision processes
- reachability analysis
- partially observable
- control system
- reinforcement learning algorithms
- finite horizon
- average reward
- risk sensitive
- reward function
- model based reinforcement learning
- optimal control
- semi markov decision processes
- markov decision process
- np hard
- control policy
- action space
- state and action spaces