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

Metric learning has become an attractive field for research on the latestyears. Loss functions like contrastive loss, triplet loss or multi-class N-pairloss have made possible generating models capable of tackling complex scenarioswith the presence of many classes and scarcity on the number of images perclass not only work to build classifiers, but to many other applications wheremeasuring similarity is the key. Deep Neural Networks trained via metriclearning also offer the possibility to solve few-shot learning problems.Currently used state of the art loss functions such as triplet and contrastiveloss functions, still suffer from slow convergence due to the selection ofeffective training samples that has been partially solved by the multi-classN-pair loss by simultaneously adding additional samples from the differentclasses. In this work, we extend triplet and multiclass-N-pair loss function byproposing the constellation loss metric where the distances among all classcombinations are simultaneously learned. We have compared our constellationloss for visual class embedding showing that our loss function over-performsthe other methods by obtaining more compact clusters while achieving betterclassification results.

Constellation Loss: Improving the efficiency of deep metric learning loss functions for optimal embedding