HARMONIA: Interpretable Graph Learning through Mixtures of Neural Bases
HARMONIA achieves stronger explanation recovery than existing interpretable graph baselines while maintaining competitive predictive performance and scaling to graphs with millions of nodes, and shows that interpretable graph learning can remain both faithful and scalable without sacrificing predictive effectiveness.
Q. D. Bùi, D. Nguyen, A. Dang et al.
· 0 citations