-regularized linear DNNs: One-way jumps from high to low rank.
Zihan WangArthur JacotPublished in: CoRR (2023)
Keyphrases
- low rank
- missing data
- linear combination
- convex optimization
- matrix factorization
- low rank matrix
- singular value decomposition
- rank minimization
- high dimensional data
- matrix completion
- matrix decomposition
- kernel matrix
- semi supervised
- high order
- low rank matrices
- singular values
- trace norm
- data matrix
- low rank approximation
- small number
- active learning
- computer vision
- low rank representation
- data sets