Optimal Shrinkage Covariance Matrix Estimation Under Random Sampling From Elliptical Distributions.
Esa OllilaElias RaninenPublished in: IEEE Trans. Signal Process. (2019)
Keyphrases
- covariance matrix
- random sampling
- sample size
- estimation error
- covariance matrices
- random samples
- normal distribution
- multivariate gaussian
- multivariate normal
- class conditional densities
- worst case
- active learning
- sampling algorithm
- hyperparameters
- mahalanobis distance
- model selection
- upper bound
- gaussian mixture
- gaussian distribution
- correlation matrix
- random sample
- sliding window
- probability distribution
- data sets
- reservoir sampling
- eigenvalues and eigenvectors
- geometrical interpretation
- optimal solution
- generalization error
- closed form
- random variables
- maximum likelihood
- multi class
- learning environment
- conjugate priors
- objective function
- feature extraction