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

Causal machine learning methods which flexibly generate heterogeneoustreatment effect estimates could be very useful tools for governments trying tomake and implement policy. However, as the critical artificial intelligenceliterature has shown, governments must be very careful of unintendedconsequences when using machine learning models. One way to try and protectagainst unintended bad outcomes is with AI Fairness methods which seek tocreate machine learning models where sensitive variables like race or gender donot influence outcomes. In this paper we argue that standard AI Fairnessapproaches developed for predictive machine learning are not suitable for allcausal machine learning applications because causal machine learning generally(at least so far) uses modelling to inform a human who is the ultimatedecision-maker while AI Fairness approaches assume a model that is makingdecisions directly. We define these scenarios as indirect and directdecision-making respectively and suggest that policy-making is best seen as ajoint decision where the causal machine learning model usually only hasindirect power. We lay out a definition of fairness for this scenario - a modelthat provides the information a decision-maker needs to accurately make a valuejudgement about just policy outcomes - and argue that the complexity of causalmachine learning models can make this difficult to achieve. The solution hereis not traditional AI Fairness adjustments, but careful modelling and awarenessof some of the decision-making biases that these methods might encourage whichwe describe.

Fairness Implications of Heterogeneous Treatment Effect Estimation with Machine Learning Methods in Policy-making