Value iteration and approximately optimal stationary policies in finite-state average Markov decision chains.
Rolando Cavazos-CadenaPublished in: Math. Methods Oper. Res. (2002)
Keyphrases
- markov decision chains
- finite state
- stationary policies
- average cost
- approximately optimal
- markov decision processes
- action sets
- markov chain
- optimal policy
- model checking
- markov decision process
- long run
- risk sensitive
- partially observable markov decision processes
- approximation ratio
- state space
- mechanism design
- multistage
- dynamic programming
- infinite horizon
- reinforcement learning
- approximation algorithms
- policy iteration
- special case
- objective function