Extended low-rank plus diagonal adaptation for deep and recurrent neural networks.
Yong ZhaoJinyu LiKshitiz KumarYifan GongPublished in: ICASSP (2017)
Keyphrases
- recurrent neural networks
- low rank
- linear combination
- convex optimization
- matrix factorization
- missing data
- neural network
- low rank matrix
- matrix completion
- singular value decomposition
- rank minimization
- feed forward
- semi supervised
- recurrent networks
- high dimensional data
- matrix decomposition
- echo state networks
- artificial neural networks
- kernel matrix
- high order
- trace norm
- low rank matrices
- robust principal component analysis
- minimization problems
- covariance matrix
- non rigid structure from motion
- learning algorithm
- negative matrix factorization
- fuzzy logic
- image processing