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

Active learning has proven to be useful for minimizing labeling costs byselecting the most informative samples. However, existing active learningmethods do not work well in realistic scenarios such as imbalance or rareclasses, out-of-distribution data in the unlabeled set, and redundancy. In thiswork, we propose SIMILAR (Submodular Information Measures based actIveLeARning), a unified active learning framework using recently proposedsubmodular information measures (SIM) as acquisition functions. We argue thatSIMILAR not only works in standard active learning, but also easily extends tothe realistic settings considered above and acts as a one-stop solution foractive learning that is scalable to large real-world datasets. Empirically, weshow that SIMILAR significantly outperforms existing active learning algorithmsby as much as ~5% - 18% in the case of rare classes and ~5% - 10% in the caseof out-of-distribution data on several image classification tasks likeCIFAR-10, MNIST, and ImageNet. SIMILAR is available as a part of the DISTILtoolkit: "https://github.com/decile-team/distil".

SIMILAR: Submodular Information Measures Based Active Learning In Realistic Scenarios