FairDLA: Improving the fairness-utility trade-off in graph neural networks via dual-level alignment

Ying Zhen, Yuchang Zhu, Jintang Li, Liang Chen

Abstract: Highlights•We observe a suboptimal fairness-utility trade-off in existing methods and investigate it via dual-level alignment, an under-explored perspective.•We propose FairDLA, a novel graph representation learning framework that achieves a better fairness-utility trade-off through dual-level alignment.•We conduct extensive experiments on five real-world datasets, the results demonstrate the superior performance of our method compared to other state-of-the-art fairness methods.

External IDs:dblp:journals/kbs/ZhenZLC25