A value-iteration scheme for undiscounted multichain Markov renewal programs.
Paul J. SchweitzerPublished in: Z. Oper. Research (1984)
Keyphrases
- markov decision processes
- average reward
- markov chain
- optimal policy
- state space
- policy iteration
- infinite horizon
- long run
- stochastic games
- reinforcement learning
- markov decision process
- finite state
- steady state
- dynamic programming
- average cost
- detection scheme
- decision processes
- conditional independence
- partially observable markov decision processes
- classification scheme
- markov process
- data sets
- markov decision problems