Insufficiently Justified Disparate Impact: A New Criterion for Subgroup Fairness
This work addresses fairness in algorithmic decision-making for domains like criminal justice and finance, offering a novel criterion to detect and mitigate unfair disparities, though it is incremental as it builds on existing fairness metrics.
The paper introduces a new fairness criterion called 'insufficiently justified disparate impact' (IJDI) to assess algorithmic recommendations, focusing on error rate imbalances between groups even after adjusting for base rates, and proposes an efficient method, IJDI-Scan, to identify unfair subpopulations, with experiments on recidivism and credit scoring data showing its effectiveness.
In this paper, we develop a new criterion, "insufficiently justified disparate impact" (IJDI), for assessing whether recommendations (binarized predictions) made by an algorithmic decision support tool are fair. Our novel, utility-based IJDI criterion evaluates false positive and false negative error rate imbalances, identifying statistically significant disparities between groups which are present even when adjusting for group-level differences in base rates. We describe a novel IJDI-Scan approach which can efficiently identify the intersectional subpopulations, defined across multiple observed attributes of the data, with the most significant IJDI. To evaluate IJDI-Scan's performance, we conduct experiments on both simulated and real-world data, including recidivism risk assessment and credit scoring. Further, we implement and evaluate approaches to mitigating IJDI for the detected subpopulations in these domains.