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/

Zendy Plus

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

Zendy Tools

Zendy Open

Semi-supervised learning (SSL) has been a powerful strategy to incorporatefew labels in learning better representations. In this paper, we focus on apractical scenario that one aims to apply SSL when unlabeled data may containout-of-class samples - those that cannot have one-hot encoded labels from aclosed-set of classes in label data, i.e., the unlabeled data is an open-set.Specifically, we introduce OpenCoS, a simple framework for handling thisrealistic semi-supervised learning scenario based upon a recent framework ofself-supervised visual representation learning. We first observe that theout-of-class samples in the open-set unlabeled dataset can be identifiedeffectively via self-supervised contrastive learning. Then, OpenCoS utilizesthis information to overcome the failure modes in the existing state-of-the-artsemi-supervised methods, by utilizing one-hot pseudo-labels and soft-labels forthe identified in- and out-of-class unlabeled data, respectively. Our extensiveexperimental results show the effectiveness of OpenCoS under the presence ofout-of-class samples, fixing up the state-of-the-art semi-supervised methods tobe suitable for diverse scenarios involving open-set unlabeled data.

OpenCoS: Contrastive Semi-supervised Learning for Handling Open-set Unlabeled Data