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 with strong and weak labelers considers a practical settingwhere we have access to both costly but accurate strong labelers and inaccuratebut cheap predictions provided by weak labelers. We study this problem in thestreaming setting, where decisions must be taken \textit{online}. We design anovel algorithmic template, Weak Labeler Active Cover (WL-AC), that is able torobustly leverage the lower quality weak labelers to reduce the querycomplexity while retaining the desired level of accuracy. Prior active learningalgorithms with access to weak labelers learn a difference classifier whichpredicts where the weak labels differ from strong labelers; this requires thestrong assumption of realizability of the difference classifier (Zhang andChaudhuri,2015). WL-AC bypasses this \textit{realizability} assumption and thusis applicable to many real-world scenarios such as random corrupted weak labelsand high dimensional family of difference classifiers (\textit{e.g.,} deepneural nets). Moreover, WL-AC cleverly trades off evaluating the quality withfull exploitation of weak labelers, which allows to convert any active learningstrategy to one that can leverage weak labelers. We provide an instantiation ofthis template that achieves the optimal query complexity for any given weaklabeler, without knowing its accuracy a-priori. Empirically, we propose aninstantiation of the WL-AC template that can be efficiently implemented forlarge-scale models (\textit{e.g}., deep neural nets) and show its effectivenesson the corrupted-MNIST dataset by significantly reducing the number of labelswhile keeping the same accuracy as in passive learning.

Improved Adaptive Algorithm for Scalable Active Learning with Weak Labeler