Fair Bayesian Data Selection via Generalized Discrepancy Measures
This addresses fairness problems in high-stakes applications by offering a scalable, data-centric solution with theoretical guarantees, though it is incremental as it builds on existing discrepancy measures.
The paper tackles fairness issues in machine learning by proposing a Bayesian data selection framework that aligns group-specific posterior distributions to mitigate biases, achieving improved fairness and accuracy on benchmark datasets.
Fairness concerns are increasingly critical as machine learning models are deployed in high-stakes applications. While existing fairness-aware methods typically intervene at the model level, they often suffer from high computational costs, limited scalability, and poor generalization. To address these challenges, we propose a Bayesian data selection framework that ensures fairness by aligning group-specific posterior distributions of model parameters and sample weights with a shared central distribution. Our framework supports flexible alignment via various distributional discrepancy measures, including Wasserstein distance, maximum mean discrepancy, and $f$-divergence, allowing geometry-aware control without imposing explicit fairness constraints. This data-centric approach mitigates group-specific biases in training data and improves fairness in downstream tasks, with theoretical guarantees. Experiments on benchmark datasets show that our method consistently outperforms existing data selection and model-based fairness methods in both fairness and accuracy.