Simple Graph Convolutional Networks

TL;DR

This paper introduces simple, theoretically grounded graph convolution operators that can be used in single-layer networks, achieving state-of-the-art results on benchmark datasets.

Contribution

It proposes new simple graph convolution operators that are easier to implement and theoretically justified, improving performance over more complex models.

Findings

State-of-the-art predictive performance on benchmark datasets

Operators are simpler and more theoretically grounded

Effective in single-layer graph convolutional networks

Abstract

Many neural networks for graphs are based on the graph convolution operator, proposed more than a decade ago. Since then, many alternative definitions have been proposed, that tend to add complexity (and non-linearity) to the model. In this paper, we follow the opposite direction by proposing simple graph convolution operators, that can be implemented in single-layer graph convolutional networks. We show that our convolution operators are more theoretically grounded than many proposals in literature, and exhibit state-of-the-art predictive performance on the considered benchmark datasets.