Average Cost Markov Decision Processes with Weakly Continuous Transition Probabilities.
Eugene A. FeinbergPavlo O. KasyanovNina V. ZadoianchukPublished in: Math. Oper. Res. (2012)
Keyphrases
- average cost
- markov decision processes
- transition probabilities
- markov chain
- action space
- finite state
- state space
- reward function
- random walk
- optimal policy
- markov decision problems
- markov models
- markov decision process
- reinforcement learning
- finite horizon
- initial state
- partially observable
- dynamic programming
- policy iteration
- infinite horizon
- long run
- action sets
- average reward
- approximate dynamic programming
- link structure
- hidden variables
- stationary policies
- risk sensitive
- maximum entropy
- special case