Strong average optimality criterion for continuous-time Markov decision processes.
Qingda WeiXian ChenPublished in: Kybernetika (2014)
Keyphrases
- markov decision processes
- optimality criterion
- average reward
- discounted reward
- average cost
- state space
- stationary policies
- optimal policy
- risk sensitive
- policy iteration
- finite state
- semi markov decision processes
- markov chain
- reinforcement learning
- dynamic programming
- transition matrices
- long run
- total reward
- factored mdps
- evaluation function
- action sets
- optimal control
- decision theoretic planning
- planning under uncertainty
- infinite horizon
- reinforcement learning algorithms
- reward function
- state variables
- markov decision process
- partially observable
- state and action spaces
- dynamical systems