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

Deep neural networks (DNNs) have shown the state-of-the-art level ofperformances in wide range of complicated tasks. In recent years, the studieshave been actively conducted to analyze the black box characteristics of DNNsand to grasp the learning behaviours, tendency, and limitations of DNNs. Inthis paper, we investigate the limitation of DNNs in image classification taskand verify it with the method inspired by cognitive psychology. Throughanalyzing the failure cases of ImageNet classification task, we hypothesizethat the DNNs do not sufficiently learn to associate related classes ofobjects. To verify how DNNs understand the relatedness between object classes,we conducted experiments on the image database provided in cognitivepsychology. We applied the ImageNet-trained DNNs to the database consisting ofpairs of related and unrelated object images to compare the featuresimilarities and determine whether the pairs match each other. In theexperiments, we observed that the DNNs show limited performance in determiningrelatedness between object classes. In addition, the DNNs present somewhatimproved performance in discovering relatedness based on similarity, but theyperform weaker in discovering relatedness based on association. Through theseexperiments, a novel analysis of learning behaviour of DNNs is provided and thelimitation which needs to be overcome is suggested.

Can Deep Neural Networks Match the Related Objects?: A Survey on ImageNet-trained Classification Models