The Power of Graph Convolutional Networks to Distinguish Random Graph Models: Short Version
arXiv:2002.05678
Abstract
Graph convolutional networks (GCNs) are a widely used method for graph representation learning. We investigate the power of GCNs, as a function of their number of layers, to distinguish between different random graph models on the basis of the embeddings of their sample graphs. In particular, the graph models that we consider arise from graphons, which are the most general possible parameterizations of infinite exchangeable graph models and which are the central objects of study in the theory of dense graph limits. We exhibit an infinite class of graphons that are well-separated in terms of cut distance and are indistinguishable by a GCN with nonlinear activation functions coming from a certain broad class if its depth is at least logarithmic in the size of the sample graph. These results theoretically match empirical observations of several prior works. Finally, we show a converse result that for pairs of graphons satisfying a degree profile separation property, a very simple GCN architecture suffices for distinguishability. To prove our results, we exploit a connection to random walks on graphs.
arXiv commentConference version of arXiv:1910.12954
References in corpus (9)
- Semi-Supervised Classification with Graph Convolutional Networks
- Simplifying Graph Convolutional Networks
- How Powerful are Graph Neural Networks?
- Rate-optimal graphon estimation
- A Consistent Histogram Estimator for Exchangeable Graph Models
- On the equivalence between graph isomorphism testing and function approximation with GNNs
- Stability and Generalization of Graph Convolutional Neural Networks
- Provable benefits of representation learning
- Fundamental Limits of Deep Graph Convolutional Networks