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

Convolutional neural networks (CNNs) have constantly achieved betterperformance over years by introducing more complex topology, and enlarging thecapacity towards deeper and wider CNNs. This makes the manual design of CNNsextremely difficult, so the automated design of CNNs has come into the researchspotlight, which has obtained CNNs that outperform manually-designed CNNs.However, the computational cost is still the bottleneck of automaticallydesigning CNNs. In this paper, inspired by transfer learning, a newevolutionary computation based framework is proposed to efficiently evolve CNNswithout compromising the classification accuracy. The proposed frameworkleverages multi-source domains, which are smaller datasets than the targetdomain datasets, to evolve a generalised CNN block only once. And then, a newstacking method is proposed to both widen and deepen the evolved block, and agrid search method is proposed to find optimal stacking solutions. Theexperimental results show the proposed method acquires good CNNs faster than 15peer competitors within less than 40 GPU-hours. Regarding the classificationaccuracy, the proposed method gains its strong competitiveness against the peercompetitors, which achieves the best error rates of 3.46%, 18.36% and 1.76% forthe CIFAR-10, CIFAR-100 and SVHN datasets, respectively.

An Efficient Evolutionary Deep Learning Framework Based on Multi-source Transfer Learning to Evolve Deep Convolutional Neural Networks