X-Layer: Building Composable Pipelined Dataflows for Low-Rank Convolutions.
Naveen VedulaReza HojabrAhmad KhonsariArrvindh ShriramanPublished in: PACT (2021)
Keyphrases
- low rank
- convex optimization
- missing data
- matrix factorization
- linear combination
- rank minimization
- low rank matrix
- matrix completion
- singular value decomposition
- kernel matrix
- trace norm
- high order
- semi supervised
- minimization problems
- low rank representation
- high dimensional data
- robust principal component analysis
- non rigid structure from motion
- singular values
- matrix decomposition
- low rank matrices
- affinity matrix
- data matrix
- neural network
- low rank approximation
- least squares
- computer vision