Fairness criteria through the lens of directed acyclic graphical models
This work addresses fairness in algorithms for researchers and practitioners, highlighting the limitations of formulaic approaches as incremental.
The paper critiques existing fairness criteria like Equalized Odds and Calibration by Group using graphical models, arguing they are misleading and should be case-specific based on the algorithm's information.
A substantial portion of the literature on fairness in algorithms proposes, analyzes, and operationalizes simple formulaic criteria for assessing fairness. Two of these criteria, Equalized Odds and Calibration by Group, have gained significant attention for their simplicity and intuitive appeal, but also for their incompatibility. This chapter provides a perspective on the meaning and consequences of these and other fairness criteria using graphical models which reveals Equalized Odds and related criteria to be ultimately misleading. An assessment of various graphical models suggests that fairness criteria should ultimately be case-specific and sensitive to the nature of the information the algorithm processes.