Towards Provably Fair Machine Learning: Bayesian Approaches For Consistent and Transparent Predictions

2026-06-10

AI Summary

This work addresses the issue of fine-grained predictive disparities in high-stakes settings, where existing machine learning models often unfairly collapse minority subgroups due to regularization. The authors propose a Fair Bayesian Classifier that, for the first time, enforces Bayesian statistical consistency and determinacy as rigorous fairness criteria across all subgroups defined by arbitrary feature intersections. The method guarantees prediction consistency on every such subgroup and proactively abstains when consistency cannot be assured, thereby preserving reliability. Built upon Bayesian inference to model the target distribution, the approach achieves provable fine-grained group fairness. Empirical evaluations on the Adult, COMPAS, and Bank Marketing datasets demonstrate zero consistency error, along with superior accuracy and multi-calibration performance compared to current baselines.

Abstract

ML classifiers deployed in high-stakes domains produce predictions whose quality varies systematically across subgroups. For granular subgroups defined by intersections of multiple features, predictions are often inconsistent with the observed data: the model's outputs contradict the evidence available for that subgroup. This problem is exacerbated by regularisation, which improves aggregate performance by collapsing small subgroups into larger groups, disproportionately affecting demographic minorities. We define two requirements for consistent prediction: determinism (identical individuals receive identical predictions) and statistical consistency (we cannot reject, at significance level alpha, the hypothesis that the predictions for a subgroup were drawn from the Bayesian optimal target distribution inferred for that subgroup). From these requirements we derive the Fair Bayesian classifier, which enforces both across every group and subgroup simultaneously and abstains whenever no consistent deterministic prediction is possible. On three benchmark datasets (Adult, COMPAS, and Bank Marketing), standard classifiers produce statistically inconsistent predictions for a substantial proportion of subgroups. Our classifier achieves zero consistency error by construction while exceeding baseline accuracy and multicalibration on every dataset tested. Statistical consistency provides a principled foundation for prediction quality with direct implications for algorithmic fairness. Minority demographics are disproportionately concentrated in small subgroups, precisely where frequentist inference is least reliable; addressing this inference problem is therefore a necessary step toward fair ML. By enforcing Bayesian consistency at the finest resolution the data supports, the our classifier demonstrates that exhaustive subgroup fairness with principled abstention is achievable in practice.

Problem

• algorithmic fairness
• statistical consistency
• subgroup fairness
• Bayesian inference
• prediction quality

Innovation

• Bayesian consistency
• Fair Bayesian classifier
• statistical consistency
• subgroup fairness
• principled abstention