High dimensional decomposition of coherent/structured matrices via sequential column/row sampling.
Mostafa RahmaniGeorge K. AtiaPublished in: ICASSP (2017)
Keyphrases
- data matrix
- binary matrix
- high dimensional
- rows and columns
- gene expression data
- binary matrices
- parameter space
- singular value decomposition
- low rank
- high dimensionality
- original data
- high dimensional data
- matrix factorization
- missing values
- dimensionality reduction
- similarity search
- low dimensional
- microarray
- sparse data
- sample size
- monte carlo
- sampling methods
- variable selection
- decomposition methods
- high dimensional problems
- nonnegative matrix factorization
- small sample size
- microarray data
- gene expression
- feature space
- random sampling
- multi dimensional
- low rank approximation
- data points
- metric space
- sampling strategy
- least squares
- feature selection
- similarity measure
- semi supervised
- sampled data
- collaborative filtering
- linear equations
- decomposition method
- structured data
- data sets