English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Convolution neural networks (CNNs) have achieved remarkable success, buttypically accompany high computation cost and numerous redundant weightparameters. To reduce the FLOPs, structure pruning is a popular approach toremove the entire hidden structures via introducing coarse-grained sparsity.Meanwhile, plentiful pruning works leverage fine-grained sparsity instead(sparsity are randomly distributed), whereas their sparse models lack specialdesigned computing library for potential speedup. In this technical report, westudy and present an efficient convolution neural network inference system toaccelerate its forward pass by utilizing the fine-grained sparsity ofcompressed CNNs. Our developed FSCNN is established based on a set ofspecialized designed sparse data structures, operators and associatedalgorithms. Experimentally, we validate that FSCNN outperforms standard deeplearning library PyTorch on popular CNN architectures such as VGG16 ifsufficiently high sparsity exhibits. However, due to the contiguity issue ofsparse operators, FSCNN is typically not comparable with highly optimized denseoperator. Therefore, coarse-grained (structured) sparsity is our recommendationfor generic model compression.

FSCNN: A Fast Sparse Convolution Neural Network Inference System