Beltrami Flow and Neural Diffusion on Graphs
Machine Learning
2021-10-19 v1 Artificial Intelligence
Machine Learning
Abstract
We propose a novel class of graph neural networks based on the discretised Beltrami flow, a non-Euclidean diffusion PDE. In our model, node features are supplemented with positional encodings derived from the graph topology and jointly evolved by the Beltrami flow, producing simultaneously continuous feature learning and topology evolution. The resulting model generalises many popular graph neural networks and achieves state-of-the-art results on several benchmarks.
Keywords
Cite
@article{arxiv.2110.09443,
title = {Beltrami Flow and Neural Diffusion on Graphs},
author = {Benjamin Paul Chamberlain and James Rowbottom and Davide Eynard and Francesco Di Giovanni and Xiaowen Dong and Michael M Bronstein},
journal= {arXiv preprint arXiv:2110.09443},
year = {2021}
}
Comments
21 pages, 5 figures. Proceedings of the Thirty-fifth Conference on Neural Information Processing Systems (NeurIPS) 2021