On the Impact of Feature Heterophily on Link Prediction with Graph Neural Networks

J Zhu, G Li, YA Yang, J Zhu, X Cui, D Koutra - arXiv preprint arXiv …, 2024 - arxiv.org
Heterophily, or the tendency of connected nodes in networks to have different class labels or
dissimilar features, has been identified as challenging for many Graph Neural Network …

Advancing Graph Neural Networks for Complex Data: A Perspective Beyond Homophily

J Zhu - 2024 - deepblue.lib.umich.edu
Graph Neural Networks (GNNs) have demonstrated significant potential in extending the
empirical success of deep learning from Euclidean spaces to non-Euclidean, graph …