Recent work has shown that a much simpler model, simple graph convolution (SGC) (Wu et al., 2019),iscompetitivewithGCNs incommon graph machine learning benchmarks.

Graph Contrastive Learning (GCL), learning the node representations by augmenting graphs, has attracted considerable attentions.

Machine learning for node classification on graphs is a prominent area driven by applications such as recommendation systems.

We study learning problems on correlated stochastic block models with two balanced communities.

Interestingly, and in a sharp departure from previous proposals, this operation on coarsened graphs is often oriented, even when the original graph is undirected.

This prevailing paradigm heavily relies on the out-of-distribution (OOD) effect to influence the model's prediction.

This paper introduces a new Subgraph GNN framework to address these issues.