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 neural networks (DNNs) have demonstrated their outperformance in variousdomains. However, it raises a social concern whether DNNs can produce reliableand fair decisions especially when they are applied to sensitive domainsinvolving valuable resource allocation, such as education, loan, andemployment. It is crucial to conduct fairness testing before DNNs are reliablydeployed to such sensitive domains, i.e., generating as many instances aspossible to uncover fairness violations. However, the existing testing methodsare still limited from three aspects: interpretability, performance, andgeneralizability. To overcome the challenges, we propose NeuronFair, a new DNNfairness testing framework that differs from previous work in several keyaspects: (1) interpretable - it quantitatively interprets DNNs' fairnessviolations for the biased decision; (2) effective - it uses the interpretationresults to guide the generation of more diverse instances in less time; (3)generic - it can handle both structured and unstructured data. Extensiveevaluations across 7 datasets and the corresponding DNNs demonstrateNeuronFair's superior performance. For instance, on structured datasets, itgenerates much more instances (~x5.84) and saves more time (with an averagespeedup of 534.56%) compared with the state-of-the-art methods. Besides, theinstances of NeuronFair can also be leveraged to improve the fairness of thebiased DNNs, which helps build more fair and trustworthy deep learning systems.

NeuronFair: Interpretable White-Box Fairness Testing through Biased Neuron Identification