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

Transfer learning using deep neural networks as feature extractors has becomeincreasingly popular over the past few years. It allows to obtainstate-of-the-art accuracy on datasets too small to train a deep neural networkon its own, and it provides cutting edge descriptors that, combined withnonparametric learning methods, allow rapid and flexible deployment ofperforming solutions in computationally restricted settings. In this paper, weare interested in showing that the features extracted using deep neuralnetworks have specific properties which can be used to improve accuracy ofdownstream nonparametric learning methods. Namely, we demonstrate that for somedistributions where information is embedded in a few coordinates, segmentingfeature vectors can lead to better accuracy. We show how this model can beapplied to real datasets by performing experiments using three mainstream deepneural network feature extractors and four databases, in vision and audio.

Improving Accuracy of Nonparametric Transfer Learning via Vector Segmentation