Computing MDP cost function for high speed networks with sample-path and quantization.
Xi-Ren CaoJunjie WangChin-Tau LeaPublished in: Commun. Inf. Syst. (2001)
Keyphrases
- sample path
- high speed networks
- cost function
- policy iteration
- asymptotic analysis
- markov decision processes
- average reward
- markov chain
- high speed
- large deviations
- distributed database systems
- optimal policy
- markov decision process
- reinforcement learning
- network management
- network conditions
- real time
- finite state
- stationary points
- lost sales
- fixed point
- temporal difference
- network resources
- bitstream
- flow control
- utility function
- traffic control
- least squares
- objective function
- database systems