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

With the remarkable success of deep learning recently, efficient networkcompression algorithms are urgently demanded for releasing the potentialcomputational power of edge devices, such as smartphones or tablets. However,optimal network pruning is a non-trivial task which mathematically is anNP-hard problem. Previous researchers explain training a pruned network asbuying a lottery ticket. In this paper, we investigate the Magnitude-BasedPruning (MBP) scheme and analyze it from a novel perspective through Fourieranalysis on the deep learning model to guide model designation. Besidesexplaining the generalization ability of MBP using Fourier transform, we alsopropose a novel two-stage pruning approach, where one stage is to obtain thetopological structure of the pruned network and the other stage is to retrainthe pruned network to recover the capacity using knowledge distillation fromlower to higher on the frequency domain. Extensive experiments on CIFAR-10 andCIFAR-100 demonstrate the superiority of our novel Fourier analysis based MBPcompared to other traditional MBP algorithms.

Win the Lottery Ticket via Fourier Analysis: Frequencies Guided Network Pruning