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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

The purpose of feature extraction on convolutional neural networks is toreuse deep representations learnt for a pre-trained model to solve a new,potentially unrelated problem. However, raw feature extraction from all layersis unfeasible given the massive size of these networks. Recently, a supervisedmethod using complexity reduction was proposed, resulting in significantimprovements in performance for transfer learning tasks. This approach firstcomputes the discriminative power of features, and then discretises them usingthresholds computed for the task. In this paper, we analyse the behaviour ofthese thresholds, with the purpose of finding a methodology for theirestimation. After a comprehensive study, we find a very strong correlationbetween problem size and threshold value, with coefficient of determinationabove 90%. These results allow us to propose a unified model for thresholdestimation, with potential application to transfer learning tasks.

Feature discriminativity estimation in CNNs for transfer learning