CUT : Combining stochastic ordering and censoring to bound steady-state rewards and first passage time.
Jean-Michel FourneauNihal PekerginSana YounèsPublished in: QEST (2007)
Keyphrases
- steady state
- markov chain
- stochastic process
- state dependent
- product form
- operating conditions
- queue length
- lower bound
- explicit expressions
- steady states
- reinforcement learning
- arrival rate
- stochastic model
- queueing networks
- monte carlo
- markov decision processes
- neural network
- stochastic processes
- transition probabilities
- markov processes