The more the merrier: exploring logical and multistage processors in credit scoring.
The widespread adoption of AI and ML methods in key decision-making contexts has spawned a great demand for fairness in these procedures. In this work we focus on the application of fair ML in the financial world through credit scoring. Our contributions are two-fold: On the one hand, we address the existent gap concerning the application of the existing methods in the literature to the case of multiple sensitive variables through the use of a new technique called logical processors (LP). On the other hand, we introduce the novel method of multistage processors (MP) to explore whether or not the combination of fairness methods can work synergistically to produce solutions with improved fairness or accuracy. Furthermore, we explore the intersection of these two lines of research compound by exploring the integration of fairness methods in the multivariate case. The results are very promising and suggest that both logical and multistage processors succeed in their respective tasks.
Palabras clave: algorithmic fairness machine learning bias mitigation multiple sensitive features multistage processor logical processor