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

As an effective way to alleviate the burden of data annotation,semi-supervised learning (SSL) provides an attractive solution due to itsability to leverage both labeled and unlabeled data to build a predictivemodel. While significant progress has been made recently, SSL algorithms areoften evaluated and developed under the assumption that the network is randomlyinitialized. This is in sharp contrast to most vision recognition systems thatare built from fine-tuning a pretrained network for better performance. Whilethe marriage of SSL and a pretrained model seems to be straightforward, recentliterature suggests that naively applying state-of-the-art SSL with apretrained model fails to unleash the full potential of training data. In thispaper, we postulate the underlying reason is that the pretrained featurerepresentation could bring a bias inherited from the source data, and the biastends to be magnified through the self-training process in a typical SSLalgorithm. To overcome this issue, we propose to use pseudo-labels from theunlabelled data to update the feature extractor that is less sensitive toincorrect labels and only allow the classifier to be trained from the labeleddata. More specifically, we progressively adjust the feature extractor toensure its induced feature distribution maintains a good class separabilityeven under strong input perturbation. Through extensive experimental studies,we show that the proposed approach achieves superior performance over existingsolutions.

Progressive Feature Adjustment for Semi-supervised Learning from Pretrained Models