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

In recent years, deep neural networks have achieved great success in thefield of computer vision. However, it is still a big challenge to deploy thesedeep models on resource-constrained embedded devices such as mobile robots,smart phones and so on. Therefore, network compression for such platforms is areasonable solution to reduce memory consumption and computation complexity. Inthis paper, a novel channel pruning method based on genetic algorithm isproposed to compress very deep Convolution Neural Networks (CNNs). Firstly, apre-trained CNN model is pruned layer by layer according to the sensitivity ofeach layer. After that, the pruned model is fine-tuned based on knowledgedistillation framework. These two improvements significantly decrease the modelredundancy with less accuracy drop. Channel selection is a combinatorialoptimization problem that has exponential solution space. In order toaccelerate the selection process, the proposed method formulates it as a searchproblem, which can be solved efficiently by genetic algorithm. Meanwhile, atwo-step approximation fitness function is designed to further improve theefficiency of genetic process. The proposed method has been verified on threebenchmark datasets with two popular CNN models: VGGNet and ResNet. On theCIFAR-100 and ImageNet datasets, our approach outperforms severalstate-of-the-art methods. On the CIFAR-10 and SVHN datasets, the pruned VGGNetachieves better performance than the original model with 8 times parameterscompression and 3 times FLOPs reduction.

A novel channel pruning method for deep neural network compression