Denumerable controlled Markov chains with strong average optimality criterion: Bounded & unbounded costs.
Rolando Cavazos-CadenaEmmanuel Fernández-GaucherandPublished in: Math. Methods Oper. Res. (1996)
Keyphrases
- markov chain
- optimality criterion
- average reward
- finite state
- average cost
- steady state
- confidence intervals
- transition probabilities
- markov model
- markov process
- evaluation function
- random walk
- markov processes
- state space
- stochastic process
- monte carlo
- stationary distribution
- transition matrix
- mdl principle
- probabilistic automata
- total cost
- partially observable markov decision processes
- information theory
- maximum entropy
- optimal policy
- higher order