An effective low-rank compression with a joint rank selection followed by a compression-friendly training.
Moonjung EoSuhyun KangWonjong RheePublished in: Neural Networks (2023)
Keyphrases
- low rank
- missing data
- matrix factorization
- image compression
- semi supervised
- frobenius norm
- linear combination
- singular value decomposition
- training set
- high dimensional data
- rank minimization
- low rank matrices
- trace norm
- singular values
- matrix completion
- kernel matrix
- matrix decomposition
- high order
- collaborative filtering
- high dimensional