Fair Densities via Boosting the Sufficient Statistics of Exponential Families

TL;DR

This paper presents a boosting algorithm for data pre-processing that enhances fairness guarantees while improving data fit, applicable to continuous features and interpretable via decision trees, with proven theoretical guarantees.

Contribution

It introduces a novel boosting-based pre-processing method that ensures fairness and data accuracy, with theoretical guarantees and applicability to continuous features.

Findings

The method guarantees a representation and statistical rate fairness.

It can be adapted for continuous domain features.

Empirical results demonstrate improved fairness and data quality.

Abstract

We introduce a boosting algorithm to pre-process data for fairness. Starting from an initial fair but inaccurate distribution, our approach shifts towards better data fitting while still ensuring a minimal fairness guarantee. To do so, it learns the sufficient statistics of an exponential family with boosting-compliant convergence. Importantly, we are able to theoretically prove that the learned distribution will have a representation rate and statistical rate data fairness guarantee. Unlike recent optimization based pre-processing methods, our approach can be easily adapted for continuous domain features. Furthermore, when the weak learners are specified to be decision trees, the sufficient statistics of the learned distribution can be examined to provide clues on sources of (un)fairness. Empirical results are present to display the quality of result on real-world data.