In this section
@inproceedings{Ayoobi:2025,
author = {Ayoobi, H and Potyka, N and Rapberger, A and Toni, F},
publisher = {Association for the Advancement of Artificial Intelligence (AAAI)},
title = {Argumentative debates for transparent bias detection},
year = {2025}
}
TY - CPAPER
AB - As the use of AI in society grows, addressing emerging biases is essential to prevent systematic disadvantages against specific groups. Several fairness notions/algorithmic methods have been proposed, but, with very few exceptions, these tend to ignore transparency. Instead, interpretability and explainability are core requirements for algorithmic fairness, even more so than for other algorithmic solutions, given the human-oriented nature of fairness. In this paper, we contribute a novel transparent method for bias detection to empower debates about the presence of bias against individuals, based on the values of protected features for the individuals and others in their neighbourhoods. Our method builds upon techniques from (formal and computational) argumentation, whereby debates result from arguing about biases within and across neighbourhoods. We evaluate our approach experimentally and demonstrate its strengths in performance against an argumentative baseline.
AU - Ayoobi,H
AU - Potyka,N
AU - Rapberger,A
AU - Toni,F
PB - Association for the Advancement of Artificial Intelligence (AAAI)
PY - 2025///
TI - Argumentative debates for transparent bias detection
ER -
