The cost of fairness in classification
This work addresses fairness in classification for machine learning practitioners, providing theoretical insights but is incremental as it builds on existing fairness measures and frameworks.
The paper tackles the problem of learning classifiers with fairness constraints by relating fairness measures to cost-sensitive risks and deriving optimal classifiers as threshold functions, showing that the tradeoff between accuracy and fairness depends on the alignment of class-probabilities for target and sensitive features.
We study the problem of learning classifiers with a fairness constraint, with three main contributions towards the goal of quantifying the problem's inherent tradeoffs. First, we relate two existing fairness measures to cost-sensitive risks. Second, we show that for cost-sensitive classification and fairness measures, the optimal classifier is an instance-dependent thresholding of the class-probability function. Third, we show how the tradeoff between accuracy and fairness is determined by the alignment between the class-probabilities for the target and sensitive features. Underpinning our analysis is a general framework that casts the problem of learning with a fairness requirement as one of minimising the difference of two statistical risks.