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

In this paper we look at popular fairness methods that use causalcounterfactuals. These methods capture the intuitive notion that a predictionis fair if it coincides with the prediction that would have been made ifsomeone's race, gender or religion were counterfactually different. In order toachieve this, we must have causal models that are able to capture what someonewould be like if we were to counterfactually change these traits. However, weargue that any model that can do this must lie outside the particularly wellbehaved class that is commonly considered in the fairness literature. This isbecause in fairness settings, models in this class entail a particularly strongcausal assumption, normally only seen in a randomised controlled trial. Weargue that in general this is unlikely to hold. Furthermore, we show in manycases it can be explicitly rejected due to the fact that samples are selectedfrom a wider population. We show this creates difficulties for counterfactualfairness as well as for the application of more general causal fairnessmethods.

Selection, Ignorability and Challenges With Causal Fairness