A note on the existence of optimal stationary policies for average Markov decision processes with countable states.
Li XiaXianping GuoXi-Ren CaoPublished in: Autom. (2023)
Keyphrases
- stationary policies
- markov decision processes
- average cost
- action sets
- state space
- dynamic programming
- initial state
- markov decision process
- finite state
- optimal policy
- reinforcement learning
- markov decision problems
- linear program
- total reward
- real time dynamic programming
- lot sizing
- infinite horizon
- reinforcement learning algorithms
- state variables
- linear programming
- sufficient conditions
- average reward
- policy iteration
- finite horizon
- partially observable
- planning under uncertainty
- state transition
- optimal control
- decision processes
- markov chain
- long run
- state transitions
- finite number
- reward function
- discounted reward