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

Modern neural networks have the capacity to overfit noisy labels frequentlyfound in real-world datasets. Although great progress has been made, existingtechniques are limited in providing theoretical guarantees for the performanceof the neural networks trained with noisy labels. Here we propose a novelapproach with strong theoretical guarantees for robust training of deepnetworks trained with noisy labels. The key idea behind our method is to selectweighted subsets (coresets) of clean data points that provide an approximatelylow-rank Jacobian matrix. We then prove that gradient descent applied to thesubsets do not overfit the noisy labels. Our extensive experiments corroborateour theory and demonstrate that deep networks trained on our subsets achieve asignificantly superior performance compared to state-of-the art, e.g., 6%increase in accuracy on CIFAR-10 with 80% noisy labels, and 7% increase inaccuracy on mini Webvision.

Coresets for Robust Training of Neural Networks against Noisy Labels