Domain Adaptation meets Individual Fairness. And they get along

TL;DR

This paper explores how enforcing individual fairness can improve model performance under distribution shifts and demonstrates that domain adaptation techniques can be adapted to promote fairness.

Contribution

It shows that individual fairness interventions can enhance out-of-distribution accuracy and that domain adaptation methods can be adapted to enforce individual fairness.

Findings

Enforcing individual fairness improves out-of-distribution accuracy.

Domain adaptation methods can be adapted to promote individual fairness.

Fairness interventions can mitigate biases caused by distributional shifts.

Abstract

Many instances of algorithmic bias are caused by distributional shifts. For example, machine learning (ML) models often perform worse on demographic groups that are underrepresented in the training data. In this paper, we leverage this connection between algorithmic fairness and distribution shifts to show that algorithmic fairness interventions can help ML models overcome distribution shifts, and that domain adaptation methods (for overcoming distribution shifts) can mitigate algorithmic biases. In particular, we show that (i) enforcing suitable notions of individual fairness (IF) can improve the out-of-distribution accuracy of ML models under the covariate shift assumption and that (ii) it is possible to adapt representation alignment methods for domain adaptation to enforce individual fairness. The former is unexpected because IF interventions were not developed with distribution shifts in mind. The latter is also unexpected because representation alignment is not a common approach in the individual fairness literature.