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

Automatic recommendation systems based on deep neural networks have becomeextremely popular during the last decade. Some of these systems can however beused for applications which are ranked as High Risk by the European Commissionin the A.I. act, as for instance for online job candidate recommendation. Whenused in the European Union, commercial AI systems for this purpose will then berequired to have to proper statistical properties with regard to potentialdiscrimination they could engender. This motivated our contribution, where wepresent a novel optimal transport strategy to mitigate undesirable algorithmicbiases in multi-class neural-network classification. Our stratey is modelagnostic and can be used on any multi-class classification neural-networkmodel. To anticipate the certification of recommendation systems using textualdata, we then used it on the Bios dataset, for which the learning task consistsin predicting the occupation of female and male individuals, based on theirLinkedIn biography. Results show that it can reduce undesired algorithmicbiases in this context to lower levels than a standard strategy.

How optimal transport can tackle gender biases in multi-class neural-network classifiers for job recommendations?