Optimal Stationary Policies in General State Space Markov Decision Chains with Finite Action Sets.
Robert K. RittLinn I. SennottPublished in: Math. Oper. Res. (1992)
Keyphrases
- stationary policies
- action sets
- state space
- average cost
- markov decision processes
- markov decision chains
- finite state
- markov decision process
- dynamic programming
- optimal policy
- reinforcement learning
- finite number
- linear program
- long run
- lot sizing
- state variables
- optimal control
- sufficient conditions
- partially observable
- special case
- markov decision problems
- action space
- markov chain
- reinforcement learning algorithms
- infinite horizon
- multistage
- planning problems
- total cost
- reward function
- initial state
- random variables
- linear programming
- expected cost
- learning agent
- search space