Sparse Tensor Core: Algorithm and Hardware Co-Design for Vector-wise Sparse Neural Networks on Modern GPUs.
Maohua ZhuTao ZhangZhenyu GuYuan XiePublished in: MICRO (2019)
Keyphrases
- neural network
- sparse matrix
- learning algorithm
- preprocessing
- dynamic programming
- genetic algorithm
- similarity measure
- optimal solution
- hardware implementation
- detection algorithm
- regularized regression
- simulated annealing
- low cost
- high dimensional
- pairwise
- objective function
- artificial neural networks
- high order
- computational complexity
- compressive sensing
- theoretical guarantees
- graphics cards
- clustering algorithm