Constrained continuous-time Markov decision processes with average criteria.
Lanlan ZhangXianping GuoPublished in: Math. Methods Oper. Res. (2008)
Keyphrases
- markov decision processes
- average cost
- state space
- stationary policies
- finite state
- optimal policy
- discounted reward
- reinforcement learning
- dynamic programming
- markov chain
- transition matrices
- infinite horizon
- policy iteration
- planning under uncertainty
- optimal control
- decision theoretic planning
- reachability analysis
- average reward
- finite horizon
- reinforcement learning algorithms
- factored mdps
- action sets
- model based reinforcement learning
- real time dynamic programming
- markov decision process
- decision processes
- optimality criterion
- partially observable
- reward function
- action space
- decision diagrams
- state and action spaces
- finite number
- monte carlo
- sufficient conditions
- linear programming
- search algorithm
- learning algorithm