Argumentative Debates for Transparent Bias Detection [Technical Report]

TL;DR

This paper introduces ABIDE, a transparent bias detection framework using structured debates based on argument graphs, enhancing interpretability and performance in identifying biases in AI systems.

Contribution

The paper presents a novel argumentative debate framework for bias detection that emphasizes transparency and interpretability, addressing limitations of existing methods.

Findings

ABIDE outperforms baseline methods in bias detection accuracy.

The debate-based approach improves transparency and interpretability.

Experimental results demonstrate the effectiveness of argument graphs in bias detection.

Abstract

As the use of AI in society grows, addressing emerging biases is essential to prevent systematic discrimination. Several bias detection methods have been proposed, but, with 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. We present ABIDE (Argumentative BIas detection by DEbate), a novel framework that structures bias detection transparently as debate, guided by an underlying argument graph as understood in (formal and computational) argumentation. The arguments are about the success chances of groups in local neighbourhoods and the significance of these neighbourhoods. We evaluate ABIDE experimentally and demonstrate its strengths in performance against an argumentative baseline.