Tracing the Propagation Path: A Flow Perspective of Representation Learning on Graphs

TL;DR

This paper introduces FlowGN, a novel graph neural network framework that models information flow from source to sink, overcoming limitations of traditional GCNs related to over-smoothing and scalability by decoupling propagation from layer depth.

Contribution

It proposes a new source-to-sink propagation perspective and a flexible, efficient framework called FlowGN that improves representation learning on graphs.

Findings

FlowGN outperforms state-of-the-art GCNs on public datasets.

Decoupling propagation from layer depth enables deeper graph models.

FlowGN is computationally efficient and adaptable to various strategies.

Abstract

Graph Convolutional Networks (GCNs) have gained significant developments in representation learning on graphs. However, current GCNs suffer from two common challenges: 1) GCNs are only effective with shallow structures; stacking multiple GCN layers will lead to over-smoothing. 2) GCNs do not scale well with large, dense graphs due to the recursive neighborhood expansion. We generalize the propagation strategies of current GCNs as a \emph{"Sink$\to$Source"} mode, which seems to be an underlying cause of the two challenges. To address these issues intrinsically, in this paper, we study the information propagation mechanism in a \emph{"Source$\to$Sink"} mode. We introduce a new concept "information flow path" that explicitly defines where information originates and how it diffuses. Then a novel framework, namely Flow Graph Network (FlowGN), is proposed to learn node representations. FlowGN is computationally efficient and flexible in propagation strategies. Moreover, FlowGN decouples the layer structure from the information propagation process, removing the interior constraint of applying deep structures in traditional GCNs. Further experiments on public datasets demonstrate the superiority of FlowGN against state-of-the-art GCNs.