Tuning Fairness by Balancing Target Labels
This addresses fairness issues in ML systems for applications like loans and job interviews, providing a practical method to mitigate bias, though it appears incremental as it builds on existing probabilistic models.
The paper tackles bias in machine learning outputs by introducing a latent target output in probabilistic models to control fairness, offering a unified framework for group fairness notions like Demographic Parity and Equality of Opportunity, and enabling stable optimization and direct fairness control without intermediate parameters.
The issue of fairness in machine learning models has recently attracted a lot of attention as ensuring it will ensure continued confidence of the general public in the deployment of machine learning systems. We focus on mitigating the harm incurred by a biased machine learning system that offers better outputs (e.g. loans, job interviews) for certain groups than for others. We show that bias in the output can naturally be controlled in probabilistic models by introducing a latent target output. This formulation has several advantages: first, it is a unified framework for several notions of group fairness such as Demographic Parity and Equality of Opportunity; second, it is expressed as a marginalisation instead of a constrained problem; and third, it allows the encoding of our knowledge of what unbiased outputs should be. Practically, the second allows us to avoid unstable constrained optimisation procedures and to reuse off-the-shelf toolboxes. The latter translates to the ability to control the level of fairness by directly varying fairness target rates. In contrast, existing approaches rely on intermediate, arguably unintuitive, control parameters such as covariance thresholds.