Orthogonal Connectivity Factorization: Interpretable Decomposition of Variability in Correlation Matrices.
Aapo HyvärinenJunichiro HirayamaVesa KiviniemiMotoaki KawanabePublished in: Neural Comput. (2016)
Keyphrases
- kronecker product
- singular value decomposition
- measurement matrix
- decomposition methods
- matrix completion
- hurwitz radon
- correlation matrix
- low rank
- correlation coefficient
- tensor factorization
- data matrix
- highly correlated
- matrix factorization
- singular values
- factorization method
- sparse matrices
- pairwise
- rows and columns
- classification rules
- non rigid structure from motion
- linear algebra
- network connectivity
- matrix representation
- connected components
- low rank matrix
- binary images
- image sequences