Breaking the Dyadic Barrier: Rethinking Fairness in Link Prediction Beyond Demographic Parity
This work addresses fairness issues in graph machine learning for applications like social recommendation, which is critical to prevent societal inequalities, though it is incremental as it builds on prior fairness definitions.
The paper tackled the problem of fairness in link prediction by showing that existing dyadic fairness definitions based on demographic parity can hide subgroup disparities and are inadequate for ranking tasks, and they proposed a new assessment framework and a lightweight post-processing method that achieved state-of-the-art fairness-utility trade-offs.
Link prediction is a fundamental task in graph machine learning with applications, ranging from social recommendation to knowledge graph completion. Fairness in this setting is critical, as biased predictions can exacerbate societal inequalities. Prior work adopts a dyadic definition of fairness, enforcing fairness through demographic parity between intra-group and inter-group link predictions. However, we show that this dyadic framing can obscure underlying disparities across subgroups, allowing systemic biases to go undetected. Moreover, we argue that demographic parity does not meet desired properties for fairness assessment in ranking-based tasks such as link prediction. We formalize the limitations of existing fairness evaluations and propose a framework that enables a more expressive assessment. Additionally, we propose a lightweight post-processing method combined with decoupled link predictors that effectively mitigates bias and achieves state-of-the-art fairness-utility trade-offs.