High Accuracy Low Precision QR Factorization and Least Square Solver on GPU with TensorCore.
Shaoshuai ZhangPanruo WuPublished in: CoRR (2019)
Keyphrases
- least squares
- singular value decomposition
- high accuracy
- low rank
- real time
- gpu implementation
- qr decomposition
- high efficiency
- graphics hardware
- dimension reduction
- graphics processing units
- rapid convergence
- graphics processors
- parallel computation
- parallel computing
- optical flow
- parallel implementation
- kronecker product
- matrix factorization
- factorization method
- tensor factorization
- pairwise
- time of flight
- tree search
- parallel programming
- feature extraction
- factorization methods
- real time rendering
- face recognition
- computer vision