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

Pruning is one of the predominant approaches for compressing deep neuralnetworks (DNNs). Lately, coresets (provable data summarizations) were leveragedfor pruning DNNs, adding the advantage of theoretical guarantees on thetrade-off between the compression rate and the approximation error. However,coresets in this domain were either data-dependent or generated underrestrictive assumptions on both the model's weights and inputs. In real-worldscenarios, such assumptions are rarely satisfied, limiting the applicability ofcoresets. To this end, we suggest a novel and robust framework for computingsuch coresets under mild assumptions on the model's weights and without anyassumption on the training data. The idea is to compute the importance of eachneuron in each layer with respect to the output of the following layer. This isachieved by a combination of L\"{o}wner ellipsoid and Caratheodory theorem. Ourmethod is simultaneously data-independent, applicable to various networks anddatasets (due to the simplified assumptions), and theoretically supported.Experimental results show that our method outperforms existing coreset basedneural pruning approaches across a wide range of networks and datasets. Forexample, our method achieved a $62\%$ compression rate on ResNet50 on ImageNetwith $1.09\%$ drop in accuracy.

Pruning Neural Networks via Coresets and Convex Geometry: Towards No Assumptions