Fairness under Graph Uncertainty: Achieving Interventional Fairness with Partially Known Causal Graphs over Clusters of Variables

TL;DR

This paper introduces a framework for achieving interventional fairness in predictions using partially known causal graphs over variable clusters, which simplifies the causal knowledge required and improves fairness-accuracy trade-offs.

Contribution

It proposes a novel learning framework leveraging cluster-level causal graphs and a scalable MMD-based discrepancy measure to attain fairness with limited causal information.

Findings

Outperforms existing methods in fairness-accuracy trade-offs

Effectively uses cluster-level causal graphs for fairness

Scalable MMD approach enhances computational efficiency

Abstract

Algorithmic decisions about individuals require predictions that are not only accurate but also fair with respect to sensitive attributes such as gender and race. Causal notions of fairness align with legal requirements, yet many methods assume access to detailed knowledge of the underlying causal graph, which is a demanding assumption in practice. We propose a learning framework that achieves interventional fairness by leveraging a causal graph over \textit{clusters of variables}, which is substantially easier to estimate than a variable-level graph. With possible \textit{adjustment cluster sets} identified from such a cluster causal graph, our framework trains a prediction model by reducing the worst-case discrepancy between interventional distributions across these sets. To this end, we develop a computationally efficient barycenter kernel maximum mean discrepancy (MMD) that scales favorably with the number of sensitive attribute values. Extensive experiments show that our framework strikes a better balance between fairness and accuracy than existing approaches, highlighting its effectiveness under limited causal graph knowledge.