Auto-Tuning of Thread Assignment for Matrix-Vector Multiplication on GPUs.
Jinwei WangXirong MaYuanping ZhuJizhou SunPublished in: IEICE Trans. Inf. Syst. (2013)
Keyphrases
- sparse matrix
- eigenvalues and eigenvectors
- matrix multiplication
- transformation matrix
- matrix representation
- symmetric matrix
- floating point
- linear algebra
- general purpose
- fine tuned
- rows and columns
- vector data
- dot product
- feature vectors
- linearly independent
- weight matrix
- fuzzy membership
- euclidean norm
- parallel processing
- feature extraction
- arithmetic operations
- fine tuning
- gpu implementation
- pseudo inverse
- parameter tuning
- low rank
- rule selection
- singular value decomposition
- vector space
- finer granularity
- database systems