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

Deep neural networks have achieved remarkable accuracy in many artificialintelligence applications, e.g. computer vision, at the cost of a large numberof parameters and high computational complexity. Weight pruning can compressDNN models by removing redundant parameters in the networks, but it bringssparsity in the weight matrix, and therefore makes the computation inefficienton GPUs. Although pruning can remove more than 80% of the weights, it actuallyhurts inference performance (speed) when running models on GPUs.  Two major problems cause this unsatisfactory performance on GPUs. First,lowering convolution onto matrix multiplication reduces data reuseopportunities and wastes memory bandwidth. Second, the sparsity brought bypruning makes the computation irregular, which leads to inefficiency whenrunning on massively parallel GPUs. To overcome these two limitations, wepropose Escort, an efficient sparse convolutional neural networks on GPUs.Instead of using the lowering method, we choose to compute the sparseconvolutions directly. We then orchestrate the parallelism and locality for thedirect sparse convolution kernel, and apply customized optimization techniquesto further improve performance. Evaluation on NVIDIA GPUs show that Escort canimprove sparse convolution speed by 2.63x and 3.07x, and inference speed by1.43x and 1.69x, compared to CUBLAS and CUSPARSE respectively.

Escoin: Efficient Sparse Convolutional Neural Network Inference on GPUs