Necessary and sufficient optimality conditions for average reward of controlled Markov chains.
Karel SladkýPublished in: Kybernetika (1973)
Keyphrases
- average reward
- markov chain
- optimality conditions
- nonlinear programming
- finite state
- lower level
- state space
- transition probabilities
- markov model
- random walk
- monte carlo
- markov decision processes
- optimal policy
- sample size
- reinforcement learning
- long run
- policy iteration
- data mining
- semidefinite programming
- higher order
- probabilistic model