A Geometrical Interpretation of Exponentially Embedded Families of Gaussian Probability Density Functions for Model Selection.
Russell CostaSteven KayPublished in: IEEE Trans. Signal Process. (2013)
Keyphrases
- model selection
- probability density function
- geometrical interpretation
- gaussian mixture
- mixture model
- covariance matrix
- density function
- sample size
- generalized gaussian
- gaussian mixture model
- cross validation
- hyperparameters
- density estimation
- parameter estimation
- posterior probability
- probability distribution
- machine learning
- model selection criteria
- em algorithm
- selection criterion
- bayesian information criterion
- unsupervised learning
- information criterion
- maximum likelihood
- expectation maximization
- gaussian process
- principal component analysis
- distance function
- gaussian distribution
- support vector
- objective function
- computer vision
- likelihood function
- bayesian methods
- feature vectors
- bayesian framework
- mean shift
- generative model