L1-norm low-rank linear approximation for accelerating deep neural networks.
Zhiqun ZhaoHengyou WangHao SunZhihai HePublished in: Neurocomputing (2020)
Keyphrases
- low rank
- linear approximation
- neural network
- trace norm
- low rank matrices
- robust principal component analysis
- linear combination
- basis functions
- convex optimization
- missing data
- matrix completion
- matrix factorization
- low rank matrix
- singular value decomposition
- semi supervised
- high dimensional data
- pattern recognition
- rank minimization
- kernel matrix
- high order
- norm minimization
- radial basis function
- manifold structure
- recommender systems
- multi task
- collaborative filtering
- pairwise
- image processing