Density estimation using a new AIC-type criterion and the EM algorithm for a linear combination of Gaussians.
Asem M. AliAly A. FaragPublished in: ICIP (2008)
Keyphrases
- linear combination
- density estimation
- em algorithm
- mixture model
- maximum likelihood
- expectation maximization
- gaussian mixture model
- mixture modeling
- finite mixture model
- probability density function
- log likelihood
- density function
- small number
- parameter estimation
- bayesian information criterion
- gaussian mixture
- generative model
- model selection
- maximum likelihood estimation
- probability density
- hyperparameters
- maximum a posteriori
- unsupervised learning
- reproducing kernel hilbert space
- mixture of gaussians
- outlier detection
- bayesian framework
- selection criterion
- probabilistic model