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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

A common heuristic in semi-supervised deep learning (SSDL) is to selectunlabelled data based on a notion of semantic similarity to the labelled data.For example, labelled images of numbers should be paired with unlabelled imagesof numbers instead of, say, unlabelled images of cars. We refer to thispractice as semantic data set matching. In this work, we demonstrate the limitsof semantic data set matching. We show that it can sometimes even degrade theperformance for a state of the art SSDL algorithm. We present and makeavailable a comprehensive simulation sandbox, called non-IID-SSDL, for stresstesting an SSDL algorithm under different degrees of distribution mismatchbetween the labelled and unlabelled data sets. In addition, we demonstrate thatsimple density based dissimilarity measures in the feature space of a genericclassifier offer a promising and more reliable quantitative matching criterionto select unlabelled data before SSDL training.

More Than Meets The Eye: Semi-supervised Learning Under Non-IID Data