Timezone: »

 
Poster
MagNet: A Neural Network for Directed Graphs
Xitong Zhang · Yixuan He · Nathan Brugnone · Michael Perlmutter · Matthew Hirn

Tue Dec 07 04:30 PM -- 06:00 PM (PST) @ Virtual
in Poster Session 2 »

The prevalence of graph-based data has spurred the rapid development of graph neural networks (GNNs) and related machine learning algorithms. Yet, despite the many datasets naturally modeled as directed graphs, including citation, website, and traffic networks, the vast majority of this research focuses on undirected graphs. In this paper, we propose MagNet, a GNN for directed graphs based on a complex Hermitian matrix known as the magnetic Laplacian. This matrix encodes undirected geometric structure in the magnitude of its entries and directional information in their phase. A charge parameter attunes spectral information to variation among directed cycles. We apply our network to a variety of directed graph node classification and link prediction tasks showing that MagNet performs well on all tasks and that its performance exceeds all other methods on a majority of such tasks. The underlying principles of MagNet are such that it can be adapted to other GNN architectures.

Keywords: Deep Learning · Machine Learning · Graph Learning

Author Information

Xitong Zhang (Michigan State University)
Yixuan He (University of Oxford)
Nathan Brugnone (Michigan State University)
Michael Perlmutter (University of California, Los Angeles)
Matthew Hirn (Michigan State University)

More from the Same Authors