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

Deep Active Learning (DAL) has been advocated as a promising method to reducelabeling costs in supervised learning. However, existing evaluations of DALmethods are based on different settings, and their results are controversial.To tackle this issue, this paper comprehensively evaluates 19 existing DALmethods in a uniform setting, including traditionalfully-\underline{s}upervised \underline{a}ctive \underline{l}earning (SAL)strategies and emerging \underline{s}emi-\underline{s}upervised\underline{a}ctive \underline{l}earning (SSAL) techniques. We have severalnon-trivial findings. First, most SAL methods cannot achieve higher accuracythan random selection. Second, semi-supervised training brings significantperformance improvement compared to pure SAL methods. Third, performing dataselection in the SSAL setting can achieve a significant and consistentperformance improvement, especially with abundant unlabeled data. Our findingsproduce the following guidance for practitioners: one should (i) apply SSALearly and (ii) collect more unlabeled data whenever possible, for better modelperformance.

An Empirical Study on the Efficacy of Deep Active Learning for Image Classification