Graph neural networks are intimately related to partial differential equations governing information diffusion on graphs. Thinking of GNNs as PDEs leads to a new broad class of graph ML methods.
O. Corby, R. Dieng, and C. Hébert. Conceptual Structures: Logical, Linguistic, and Computational
Issues, 8th International Conference on Conceptual Structures,
ICCS 2000, Darmstadt, Germany, August 14-18, 2000, Proceedings, volume 1867 of LNCS, page 468--482. Springer, (2000)