Fair inference on error-prone outcomes
This addresses fairness issues in ML for domains like healthcare where data is noisy, but it is incremental as it integrates existing methods.
The paper tackles the problem of unfairness in supervised learning when target labels are error-prone, showing that existing fairness methods fail for the true target variable, and proposes a framework combining fair ML and statistical measurement models to address this, demonstrating in a healthcare example that accounting for measurement error removes previously detected unfairness.
Fair inference in supervised learning is an important and active area of research, yielding a range of useful methods to assess and account for fairness criteria when predicting ground truth targets. As shown in recent work, however, when target labels are error-prone, potential prediction unfairness can arise from measurement error. In this paper, we show that, when an error-prone proxy target is used, existing methods to assess and calibrate fairness criteria do not extend to the true target variable of interest. To remedy this problem, we suggest a framework resulting from the combination of two existing literatures: fair ML methods, such as those found in the counterfactual fairness literature on the one hand, and, on the other, measurement models found in the statistical literature. We discuss these approaches and their connection resulting in our framework. In a healthcare decision problem, we find that using a latent variable model to account for measurement error removes the unfairness detected previously.