Efficient computation of normalized maximum likelihood coding for Gaussian mixtures with its applications to optimal clustering.
So HiraiKenji YamanishiPublished in: ISIT (2011)
Keyphrases
- efficient computation
- normalized maximum likelihood
- gaussian mixture
- change point detection
- spatially adaptive
- minimum description length
- closed form
- outlier detection
- gaussian mixture model
- computational efficiency
- mdl principle
- density estimation
- covariance matrix
- denoising
- sequential data
- probability distribution
- data sets
- data objects
- non stationary
- mixture model
- em algorithm
- natural images
- dimensionality reduction
- training set