Fair Bayes-Optimal Classifiers Under Predictive Parity
This addresses fairness concerns in AI by focusing on the understudied sufficiency-based measure of predictive parity, though it is incremental as it builds on existing fair classification frameworks.
The paper tackles the problem of achieving predictive parity in fair machine learning by proving that fair Bayes-optimal classifiers under this condition are group-wise thresholding rules when group performances vary moderately, and proposes the FairBayes-DPP algorithm to enforce predictive parity while maximizing accuracy.
Increasing concerns about disparate effects of AI have motivated a great deal of work on fair machine learning. Existing works mainly focus on independence- and separation-based measures (e.g., demographic parity, equality of opportunity, equalized odds), while sufficiency-based measures such as predictive parity are much less studied. This paper considers predictive parity, which requires equalizing the probability of success given a positive prediction among different protected groups. We prove that, if the overall performances of different groups vary only moderately, all fair Bayes-optimal classifiers under predictive parity are group-wise thresholding rules. Perhaps surprisingly, this may not hold if group performance levels vary widely; in this case we find that predictive parity among protected groups may lead to within-group unfairness. We then propose an algorithm we call FairBayes-DPP, aiming to ensure predictive parity when our condition is satisfied. FairBayes-DPP is an adaptive thresholding algorithm that aims to achieve predictive parity, while also seeking to maximize test accuracy. We provide supporting experiments conducted on synthetic and empirical data.