Averaging, maximum penalized likelihood and Bayesian estimation for improving Gaussian mixture probability density estimates.
Dirk OrmoneitVolker TrespPublished in: IEEE Trans. Neural Networks (1998)
Keyphrases
- bayesian estimation
- gaussian mixture
- probability density
- density estimation
- em algorithm
- gaussian mixture model
- posterior distribution
- probability distribution
- expectation maximization
- probability density function
- log likelihood
- closed form
- covariance matrix
- hidden variables
- mixture model
- density function
- posterior probability
- maximum likelihood
- outlier detection
- maximum likelihood estimation
- random variables
- computer vision
- generative model
- least squares
- feature vectors