Accelerating sparse matrix-matrix multiplication with GPU Tensor Cores.
Orestis ZachariadisNitin SatputeJuan Gómez-LunaJoaquín OlivaresPublished in: Comput. Electr. Eng. (2020)
Keyphrases
- sparse matrix
- matrix multiplication
- message passing
- distributed memory
- high order
- parallel implementation
- higher order
- matrix factorization
- floating point
- diffusion tensor
- parallel computing
- graphics processing units
- dimensionality reduction
- random projections
- gpu implementation
- belief propagation
- markov random field
- bayesian networks