A self-calibrated direct approach to precision matrix estimation and linear discriminant analysis in high dimensions.
Chi Seng PunMatthew Zakharia HadimajaPublished in: Comput. Stat. Data Anal. (2021)
Keyphrases
- linear discriminant analysis
- high dimensions
- high dimensional data
- null space
- dimensionality reduction
- low rank
- discriminant analysis
- qr decomposition
- high dimensional
- small sample size
- dimension reduction
- low dimensional
- face recognition
- high dimensional spaces
- singular value decomposition
- data points
- similarity search
- data sets
- data analysis
- nearest neighbor
- principal component analysis
- support vector machine svm
- scatter matrices
- feature extraction
- support vector
- high dimensionality
- fisher criterion
- decomposition method
- poor quality
- discriminative information
- sparse representation
- intrinsic dimensionality
- projection matrix
- feature space
- scatter matrix
- low quality
- dimensional data
- euclidean distance