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

Adversarial training has been shown to be reliable in improving robustnessagainst adversarial samples. However, the problem of adversarial training interms of fairness has not yet been properly studied, and the relationshipbetween fairness and accuracy attack still remains unclear. Can wesimultaneously improve robustness w.r.t. both fairness and accuracy? To tacklethis topic, in this paper, we study the problem of adversarial training andadversarial attack w.r.t. both metrics. We propose a unified structure forfairness attack which brings together common notions in group fairness, and wetheoretically prove the equivalence of fairness attack against differentnotions. Moreover, we show the alignment of fairness and accuracy attack, andtheoretically demonstrate that robustness w.r.t. one metric benefits fromrobustness w.r.t. the other metric. Our study suggests a novel way to unifyadversarial training and attack w.r.t. fairness and accuracy, and experimentalresults show that our proposed method achieves better performance in terms ofrobustness w.r.t. both metrics.

To be Robust and to be Fair: Aligning Fairness with Robustness