Determining the optimal strategies for discrete control problems on stochastic networks with discounted costs.
Dmitrii LozovanuStefan PicklPublished in: Discret. Appl. Math. (2015)
Keyphrases
- control problems
- optimal strategy
- continuous state spaces
- expected cost
- optimal control
- monte carlo
- reinforcement learning
- stochastic control
- decision problems
- average cost
- markov decision processes
- infinite horizon
- adaptive control
- optimal policy
- state space
- queueing systems
- expected utility
- markov processes
- dynamic programming
- mathematical models
- markov decision process
- total cost
- machine learning
- long run average cost
- finite horizon
- steady state
- decision makers
- np hard
- artificial intelligence