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

Deep neural networks (DNNs) have great expressive power, which can evenmemorize samples with wrong labels. It is vitally important to reiteraterobustness and generalization in DNNs against label corruption. To this end,this paper studies the 0-1 loss, which has a monotonic relationship with anempirical adversary (reweighted) risk~\citep{hu2016does}. Although the 0-1 losshas some robust properties, it is difficult to optimize. To efficientlyoptimize the 0-1 loss while keeping its robust properties, we propose a verysimple and efficient loss, i.e. curriculum loss (CL). Our CL is a tighter upperbound of the 0-1 loss compared with conventional summation based surrogatelosses. Moreover, CL can adaptively select samples for model training. As aresult, our loss can be deemed as a novel perspective of curriculum sampleselection strategy, which bridges a connection between curriculum learning androbust learning. Experimental results on benchmark datasets validate therobustness of the proposed loss.

Curriculum Loss: Robust Learning and Generalization against Label Corruption