Repairing without Retraining: Avoiding Disparate Impact with Counterfactual Distributions
This addresses fairness issues in machine learning for applications where retraining models is impractical, though it is incremental as it builds on existing distribution-based fairness methods.
The paper tackles the problem of performance disparities in black-box classifiers across groups defined by sensitive attributes by perturbing input distributions for disadvantaged groups, resulting in reduced disparate impact without significant accuracy loss in experiments on real-world datasets.
When the performance of a machine learning model varies over groups defined by sensitive attributes (e.g., gender or ethnicity), the performance disparity can be expressed in terms of the probability distributions of the input and output variables over each group. In this paper, we exploit this fact to reduce the disparate impact of a fixed classification model over a population of interest. Given a black-box classifier, we aim to eliminate the performance gap by perturbing the distribution of input variables for the disadvantaged group. We refer to the perturbed distribution as a counterfactual distribution, and characterize its properties for common fairness criteria. We introduce a descent algorithm to learn a counterfactual distribution from data. We then discuss how the estimated distribution can be used to build a data preprocessor that can reduce disparate impact without training a new model. We validate our approach through experiments on real-world datasets, showing that it can repair different forms of disparity without a significant drop in accuracy.