Mutual Information Maximization in Graph Neural Networks
This work addresses information loss in GNNs for graph representation learning, offering incremental improvements over existing methods.
The paper tackled the problem of information loss in graph neural networks (GNNs) during aggregation and iteration by proposing an approach to enlarge the normal neighborhood to maximize mutual information, resulting in improved state-of-the-art performance on four graph tasks including supervised and semi-supervised classification, link prediction, and edge generation.
A variety of graph neural networks (GNNs) frameworks for representation learning on graphs have been recently developed. These frameworks rely on aggregation and iteration scheme to learn the representation of nodes. However, information between nodes is inevitably lost in the scheme during learning. In order to reduce the loss, we extend the GNNs frameworks by exploring the aggregation and iteration scheme in the methodology of mutual information. We propose a new approach of enlarging the normal neighborhood in the aggregation of GNNs, which aims at maximizing mutual information. Based on a series of experiments conducted on several benchmark datasets, we show that the proposed approach improves the state-of-the-art performance for four types of graph tasks, including supervised and semi-supervised graph classification, graph link prediction and graph edge generation and classification.