Detection and Mitigation of Algorithmic Bias via Predictive Rate Parity

TL;DR

This paper introduces a new non-parametric test for predictive parity in dependent data settings and proposes a mitigation method to reduce bias, addressing limitations of calibration-based approaches in large-scale applications.

Contribution

It develops a statistically rigorous test for predictive parity with dependent observations and offers a mitigation strategy to improve fairness in model outputs.

Findings

Existing calibration-based tests are inadequate for dependent data.

The proposed test significantly differs under various data assumptions.

The mitigation method effectively reduces bias in model predictions.

Abstract

Predictive parity (PP), also known as sufficiency, is a core definition of algorithmic fairness essentially stating that model outputs must have the same interpretation of expected outcomes regardless of group. Testing and satisfying PP is especially important in many settings where model scores are interpreted by humans or directly provide access to opportunity, such as healthcare or banking. Solutions for PP violations have primarily been studied through the lens of model calibration. However, we find that existing calibration-based tests and mitigation methods are designed for independent data, which is often not assumable in large-scale applications such as social media or medical testing. In this work, we address this issue by developing a statistically rigorous non-parametric regression based test for PP with dependent observations. We then apply our test to illustrate that PP testing can significantly vary under the two assumptions. Lastly, we provide a mitigation solution to provide a minimally-biased post-processing transformation function to achieve PP.