Privilege Scores

MCML Authors

Ludwig Bothmann

Dr.

→ Group Bernd Bischl
 Statistical Learning and Data Science

Philip Amir Boustani

→ Group Bernd Bischl
 Statistical Learning and Data Science

Giuseppe Casalicchio

Dr.

→ Group Bernd Bischl
 Statistical Learning and Data Science

Bernd Bischl

Prof. Dr.

Director

Statistical Learning and Data Science

Abstract

Bias-transforming methods of fairness-aware machine learning aim to correct a non-neutral status quo with respect to a protected attribute (PA). Current methods, however, lack an explicit formulation of what drives non-neutrality. We introduce privilege scores (PS) to measure PA-related privilege by comparing the model predictions in the real world with those in a fair world in which the influence of the PA is removed. At the individual level, PS can identify individuals who qualify for affirmative action; at the global level, PS can inform bias-transforming policies. After presenting estimation methods for PS, we propose privilege score contributions (PSCs), an interpretation method that attributes the origin of privilege to mediating features and direct effects. We provide confidence intervals for both PS and PSCs. Experiments on simulated and real-world data demonstrate the broad applicability of our methods and provide novel insights into gender and racial privilege in mortgage and college admissions applications.

inproceedings

---

EWAF 2025

4th European Workshop on Algorithmic Fairness. Eindhoven, The Netherlands, Jun 30-Jul 02, 2025.

---

Authors

L. Bothmann • P. A. Boustani • J. M. Alvarez • G. Casalicchio • B. Bischl • S. Dandl

---

Links

---

Research Area

 A1 | Statistical Foundations & Explainability