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

Acoustic Scene Classification (ASC) algorithms are usually expected to bedeployed in resource-constrained systems. Existing works reduce the complexityof ASC algorithms by pruning some components, e.g. pruning channels in neuralnetwork. In practice, neural networks are often trained with sparsificationsuch that unimportant channels can be found and further pruned. However, littleefforts have been made to explore the the impact of channel sparsity on neuralnetwork pruning. To fully utilize the benefits of pruning for ASC, and to makesure the model performs consistently, we need a more profound comprehension ofchannel sparsification and its effects. This paper examines the internalweights acquired by convolutional neural networks that will undergone pruning.The study discusses how these weights can be utilized to create a novel metric,Weight Skewness (WS), for quantifying the sparsity of channels. We also providea new approach to compare the performance of different pruning methods, whichbalances the trade-off between accuracy and complexity. The experiment resultsdemonstrate that 1) applying higher channel sparsity to models can achievegreater compression rates while maintaining acceptable levels of accuracy; 2)the selection of pruning method has little influence on result 1); 3)MobileNets exhibit more significant benefits from channel sparsification thanVGGNets and ResNets.

Exploring the Effects of Channel Sparsity on Neural Network Pruning for Acoustic Scene Classification