The EM Algorithm is Adaptively-Optimal for Unbalanced Symmetric Gaussian Mixtures.
Nir WeinbergerGuy BreslerPublished in: CoRR (2021)
Keyphrases
- em algorithm
- gaussian mixture
- expectation maximization
- mixture model
- closed form
- maximum likelihood
- parameter estimation
- gaussian mixture model
- generative model
- hyperparameters
- maximum likelihood estimation
- probability density function
- log likelihood
- mixture of gaussians
- model based clustering
- maximum a posteriori
- mixture modeling
- penalized likelihood
- density estimation
- bayesian framework
- worst case
- covariance matrix
- unsupervised learning
- knowledge discovery
- probabilistic model
- k means
- expectation maximisation
- image segmentation