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

It is well-known that a deep neural network has a strong fitting capabilityand can easily achieve a low training error even with randomly assigned classlabels. When the number of training samples is small, or the class labels arenoisy, networks tend to memorize patterns specific to individual instances tominimize the training error. This leads to the issue of overfitting and poorgeneralisation performance. This paper explores a remedy by suppressing thenetwork's tendency to rely on instance-specific patterns for empirical errorminimisation. The proposed method is based on an adversarial trainingframework. It suppresses features that can be utilized to identify individualinstances among samples within each class. This leads to classifiers only usingfeatures that are both discriminative across classes and common within eachclass. We call our method Adversarial Suppression of Identity Features (ASIF),and demonstrate the usefulness of this technique in boosting generalisationaccuracy when faced with small datasets or noisy labels. Our source code isavailable.

Regularizing Neural Network Training via Identity-wise Discriminative Feature Suppression