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

The recently proposed Temporal Ensembling has achieved state-of-the-artresults in several semi-supervised learning benchmarks. It maintains anexponential moving average of label predictions on each training example, andpenalizes predictions that are inconsistent with this target. However, becausethe targets change only once per epoch, Temporal Ensembling becomes unwieldywhen learning large datasets. To overcome this problem, we propose MeanTeacher, a method that averages model weights instead of label predictions. Asan additional benefit, Mean Teacher improves test accuracy and enables trainingwith fewer labels than Temporal Ensembling. Without changing the networkarchitecture, Mean Teacher achieves an error rate of 4.35% on SVHN with 250labels, outperforming Temporal Ensembling trained with 1000 labels. We alsoshow that a good network architecture is crucial to performance. Combining MeanTeacher and Residual Networks, we improve the state of the art on CIFAR-10 with4000 labels from 10.55% to 6.28%, and on ImageNet 2012 with 10% of the labelsfrom 35.24% to 9.11%.

Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results