Covariance Matrix Estimation with Multi-Regularization Parameters based on MDL Principle.
Xiuling ZhouPing GuoC. L. Philip ChenPublished in: Neural Process. Lett. (2013)
Keyphrases
- covariance matrix
- mdl principle
- regularization parameter
- generalized cross validation
- estimation error
- covariance matrices
- sample size
- principal component analysis
- image restoration
- noise variance
- information theory
- hyperparameters
- cross validation
- geometrical interpretation
- maximum likelihood
- evolutionary algorithm
- noise level
- correlation matrix
- principal components
- blur identification
- model selection
- feature extraction
- parameter estimation
- maximum likelihood estimation
- noisy images
- objective function
- machine learning