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

Given data with label noise (i.e., incorrect data), deep neural networkswould gradually memorize the label noise and impair model performance. Torelieve this issue, curriculum learning is proposed to improve modelperformance and generalization by ordering training samples in a meaningful(e.g., easy to hard) sequence. Previous work takes incorrect samples as generichard ones without discriminating between hard samples (i.e., hard samples incorrect data) and incorrect samples. Indeed, a model should learn from hardsamples to promote generalization rather than overfit to incorrect ones. Inthis paper, we address this problem by appending a novel loss functionDiscrimLoss, on top of the existing task loss. Its main effect is toautomatically and stably estimate the importance of easy samples and difficultsamples (including hard and incorrect samples) at the early stages of trainingto improve the model performance. Then, during the following stages,DiscrimLoss is dedicated to discriminating between hard and incorrect samplesto improve the model generalization. Such a training strategy can be formulateddynamically in a self-supervised manner, effectively mimicking the mainprinciple of curriculum learning. Experiments on image classification, imageregression, text sequence regression, and event relation reasoning demonstratethe versatility and effectiveness of our method, particularly in the presenceof diversified noise levels.

DiscrimLoss: A Universal Loss for Hard Samples and Incorrect Samples Discrimination