Sporadic Overtaking Optimality in Markov Decision Problems.
János FleschArkadi PredtetchinskiEilon SolanPublished in: Dyn. Games Appl. (2017)
Keyphrases
- markov decision problems
- average cost
- linear programming
- state space
- markov decision processes
- optimal policy
- partially observable
- reinforcement learning
- optimal solution
- utility function
- expected utility
- long run
- decision processes
- finite number
- transition probabilities
- policy iteration
- queueing networks
- decision theoretic
- infinite horizon
- optimal control
- linear program
- finite state
- multi agent systems
- np hard
- supervised learning
- total cost
- function approximators
- function approximation
- machine learning