Networked Inequality: Preferential Attachment Bias in Graph Neural Network Link Prediction

[PDF] arxiv.org

Theoretical and Empirical Insights into the Origins of Degree Bias in Graph Neural Networks

A Subramonian, J Kang, Y Sun - arXiv preprint arXiv:2404.03139, 2024 - arxiv.org
Graph Neural Networks (GNNs) often perform better for high-degree nodes than low-degree
nodes on node classification tasks. This degree bias can reinforce social marginalization by …
被引用次数：3 相关文章 所有 2 个版本