Efficient Implementations of the EM-Algorithm for Transient Markovian Arrival Processes.
Mindaugas BrazenasGábor HorváthMiklós TelekPublished in: ASMTA (2016)
Keyphrases
- em algorithm
- arrival processes
- steady state
- queue length
- expectation maximization
- service times
- mixture model
- maximum likelihood
- arrival process
- maximum likelihood estimation
- gaussian mixture model
- generative model
- parameter estimation
- single server
- gaussian mixture
- incomplete data
- markov chain
- hyperparameters
- log likelihood
- likelihood function
- probability density function
- heavy traffic
- hidden variables
- arrival rate
- queueing networks
- queueing model
- mixture modeling
- expectation maximisation
- penalized likelihood
- density estimation
- maximum a posteriori
- unsupervised learning