Convolutional Learning on Multigraphs
This addresses the need for better modeling of intricate data structures in applications like wireless networks and social media analysis, representing an incremental advancement over existing graph neural networks.
The paper tackles the problem of modeling complex data structures where traditional graphs are insufficient by developing convolutional multigraph neural networks (MGNNs) for information processing on multigraphs, achieving improved performance in wireless resource allocation and hate speech localization tasks over traditional graph neural networks.
Graph convolutional learning has led to many exciting discoveries in diverse areas. However, in some applications, traditional graphs are insufficient to capture the structure and intricacies of the data. In such scenarios, multigraphs arise naturally as discrete structures in which complex dynamics can be embedded. In this paper, we develop convolutional information processing on multigraphs and introduce convolutional multigraph neural networks (MGNNs). To capture the complex dynamics of information diffusion within and across each of the multigraph's classes of edges, we formalize a convolutional signal processing model, defining the notions of signals, filtering, and frequency representations on multigraphs. Leveraging this model, we develop a multigraph learning architecture, including a sampling procedure to reduce computational complexity. The introduced architecture is applied towards optimal wireless resource allocation and a hate speech localization task, offering improved performance over traditional graph neural networks.