Markov Decision Processes with Incomplete Information and Semiuniform Feller Transition Probabilities.
Eugene A. FeinbergPavlo O. KasyanovMichael Z. ZgurovskyPublished in: SIAM J. Control. Optim. (2022)
Keyphrases
- incomplete information
- transition probabilities
- markov decision processes
- markov chain
- state space
- random walk
- markov models
- finite state
- optimal policy
- dynamic programming
- policy iteration
- markov decision process
- reinforcement learning
- average cost
- autonomous agents
- reinforcement learning algorithms
- nash equilibria
- markov decision problems
- first order logic
- reward function
- infinite horizon
- action space
- average reward
- long run
- maximum entropy
- hidden markov models
- link structure
- markov model
- least squares