Improving the vector (ϵ) acceleration for the EM algorithm using a re-starting procedure.
Masahiro KurodaZhi GengMichio SakakiharaPublished in: Comput. Stat. (2015)
Keyphrases
- em algorithm
- expectation maximization
- mixture model
- maximum likelihood
- gaussian mixture model
- parameter estimation
- maximum likelihood estimation
- expectation maximisation
- generative model
- incomplete data
- gaussian mixture
- hidden variables
- likelihood function
- likelihood maximization
- latent dirichlet
- sum product algorithm
- probability density function
- maximum a posteriori
- density estimation
- log likelihood
- mixture distribution
- probabilistic model
- parameter learning
- maximum likelihood estimates
- mixture modeling
- vector space
- penalized likelihood
- image segmentation