EM Algorithm with PIP Initialization and Temperature-Based Selection.
Yuta IshikawaRyohei NakanoPublished in: KES (3) (2008)
Keyphrases
- em algorithm
- expectation maximization
- mixture model
- maximum likelihood
- parameter estimation
- likelihood function
- gaussian mixture model
- expectation maximisation
- generative model
- maximum likelihood estimation
- gaussian mixture
- log likelihood
- k means
- incomplete data
- maximum a posteriori
- likelihood maximization
- hidden variables
- gibbs sampling
- penalized likelihood
- density estimation
- probabilistic model
- model based clustering
- probability density function
- bayesian framework
- parameter learning
- finite mixture model
- image processing
- finite mixtures
- sum product algorithm