Decomposing Direct and Indirect Biases in Linear Models Under Demographic Parity Constraint (Student Abstract)

Bertille Tierny; Arthur Charpentier; Francois Hu

AAAI 2026technical0 citations

Decomposing Direct and Indirect Biases in Linear Models Under Demographic Parity Constraint (Student Abstract)

Bertille Tierny, Arthur Charpentier, Francois Hu

Abstract

Linear models are widely used in high-stakes decision-making due to their interpretability, but fairness constraints like Demographic Parity (DP) create opaque effects on model coefficients and predictive bias distribution. We propose a post-processing framework that can be applied on top of any linear model to decompose bias into direct (sensitive-attribute) and indirect (correlated-features) components. Our method analytically characterizes how DP reshapes each coefficient, enabling transparent feature-level interpretation.

BibTeX

@inproceedings{aaai2026_decomposingdirec,
  title = {Decomposing Direct and Indirect Biases in Linear Models Under Demographic Parity Constraint (Student Abstract)},
  author = {Bertille Tierny and Arthur Charpentier and Francois Hu},
  booktitle = {AAAI 2026},
  year = {2026}
}