Re-compact: Structured Pruning and SpMM Kernel Co-design for Accelerating DNNs on GPUs.
Yuling ZhangAo RenXianzhang ChenQiu LinYujuan TanDuo LiuPublished in: ICCD (2023)
Keyphrases
- kernel function
- search space
- parallel processing
- pruning method
- general purpose
- kernel methods
- kernel regression
- graphics hardware
- pruning methods
- structured representation
- pruning algorithms
- component analysis
- sparse kernel
- tree pruning
- pairwise
- kernel space
- convolution kernel
- compact representations
- binary codes
- pruning strategy
- graphics processing units
- kernel matrix
- multi class