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

We show that deep networks trained to satisfy demographic parity often do sothrough a form of race or gender awareness, and that the more we force anetwork to be fair, the more accurately we can recover race or gender from theinternal state of the network. Based on this observation, we investigate analternative fairness approach: we add a second classification head to thenetwork to explicitly predict the protected attribute (such as race or gender)alongside the original task. After training the two-headed network, we enforcedemographic parity by merging the two heads, creating a network with the samearchitecture as the original network. We establish a close relationship betweenexisting approaches and our approach by showing (1) that the decisions of afair classifier are well-approximated by our approach, and (2) that an unfairand optimally accurate classifier can be recovered from a fair classifier andour second head predicting the protected attribute. We use our explicitformulation to argue that the existing fairness approaches, just as ours,demonstrate disparate treatment and that they are likely to be unlawful in awide range of scenarios under US law.

Are Two Heads the Same as One? Identifying Disparate Treatment in Fair Neural Networks