Accelerating Sparse Matrix Operations in Neural Networks on Graphics Processing Units.
Arturo ArguetaDavid ChiangPublished in: ACL (1) (2019)
Keyphrases
- sparse matrix
- neural network
- graphics processing units
- floating point
- processing units
- general purpose
- rows and columns
- gpu implementation
- pattern recognition
- parallel implementation
- computing systems
- fixed point
- efficient implementation
- random projections
- parallel processing
- real time
- parallel computing
- massively parallel
- high performance computing
- special case