A probabilistic analysis of bias optimality in unichain Markov decision processes.
Mark E. LewisMartin L. PutermanPublished in: IEEE Trans. Autom. Control. (2001)
Keyphrases
- markov decision processes
- average cost
- finite state
- state space
- reinforcement learning
- infinite horizon
- transition matrices
- long run
- optimal policy
- policy iteration
- decision processes
- planning under uncertainty
- stationary policies
- dynamic programming
- model checking
- least squares
- finite horizon
- probabilistic planning
- decision theoretic planning