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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In this paper, we propose a novel progressive parameter pruning method forConvolutional Neural Network acceleration, named Structured ProbabilisticPruning (SPP), which effectively prunes weights of convolutional layers in aprobabilistic manner. Unlike existing deterministic pruning approaches, whereunimportant weights are permanently eliminated, SPP introduces a pruningprobability for each weight, and pruning is guided by sampling from the pruningprobabilities. A mechanism is designed to increase and decrease pruningprobabilities based on importance criteria in the training process. Experimentsshow that, with 4x speedup, SPP can accelerate AlexNet with only 0.3% loss oftop-5 accuracy and VGG-16 with 0.8% loss of top-5 accuracy in ImageNetclassification. Moreover, SPP can be directly applied to acceleratemulti-branch CNN networks, such as ResNet, without specific adaptations. Our 2xspeedup ResNet-50 only suffers 0.8% loss of top-5 accuracy on ImageNet. Wefurther show the effectiveness of SPP on transfer learning tasks.

Structured Probabilistic Pruning for Convolutional Neural Network Acceleration