Learning traffic correlations in multi-class queueing systems by sampling queue lengths, with routing applications.
Martin ZubeldiaMichel MandjesPublished in: Perform. Evaluation (2021)
Keyphrases
- multi class
- queueing systems
- queue length
- queueing networks
- arrival rate
- model selection
- learning algorithm
- single server
- support vector
- heavy traffic
- steady state
- machine learning
- multi class classification
- error correcting output codes
- neural network
- stationary distribution
- multi class classifier
- cost sensitive
- learning problems
- random walk
- support vector machine
- pairwise
- similarity measure
- feature selection