The Fragility of Fairness: Causal Sensitivity Analysis for Fair Machine Learning

TL;DR

This paper introduces a causal sensitivity analysis framework to evaluate the robustness of fairness metrics in machine learning, revealing their vulnerability to dataset biases and emphasizing the need for more reliable fairness assessments.

Contribution

It adapts causal sensitivity analysis to fairness metrics, enabling comprehensive bias analysis and domain-specific constraint encoding in fair ML evaluations.

Findings

Fairness assessments are highly sensitive to dataset biases.

The framework reveals fragility of common parity metrics.

Analysis across datasets shows significant variability in fairness evaluations.

Abstract

Fairness metrics are a core tool in the fair machine learning literature (FairML), used to determine that ML models are, in some sense, "fair". Real-world data, however, are typically plagued by various measurement biases and other violated assumptions, which can render fairness assessments meaningless. We adapt tools from causal sensitivity analysis to the FairML context, providing a general framework which (1) accommodates effectively any combination of fairness metric and bias that can be posed in the "oblivious setting"; (2) allows researchers to investigate combinations of biases, resulting in non-linear sensitivity; and (3) enables flexible encoding of domain-specific constraints and assumptions. Employing this framework, we analyze the sensitivity of the most common parity metrics under 3 varieties of classifier across 14 canonical fairness datasets. Our analysis reveals the striking fragility of fairness assessments to even minor dataset biases. We show that causal sensitivity analysis provides a powerful and necessary toolkit for gauging the informativeness of parity metric evaluations. Our repository is available here: https://github.com/Jakefawkes/fragile_fair.