Regularized Computation of Approximate Pseudoinverse of Matrices Using Low-Rank Tensor Train Decompositions.
Namgil LeeAndrzej CichockiPublished in: CoRR (2015)
Keyphrases
- low rank
- pseudo inverse
- trace norm
- singular value decomposition
- frobenius norm
- reconstruction error
- least squares
- matrix completion
- matrix decomposition
- low rank matrix
- high order
- tensor decomposition
- covariance matrix
- ridge regression
- matrix factorization
- singular values
- missing data
- data matrix
- linear combination
- convex optimization
- principal component analysis
- low rank approximation
- kernel matrix
- dimensionality reduction
- semi supervised
- tensor factorization
- high dimensional data
- multi task
- norm minimization
- radial basis function
- regression model
- missing values
- manifold structure
- input data
- sparse representation
- data sets
- pattern recognition
- feature selection
- machine learning