GTNet: A Tree-Based Deep Graph Learning Architecture

TL;DR

This paper introduces GTNets, a novel deep graph learning architecture inspired by tree structures, which effectively propagates messages and avoids over-smoothing, achieving state-of-the-art results.

Contribution

The paper presents a new message passing scheme based on tree representations and proposes two models, GTAN and GTCN, that outperform existing methods.

Findings

State-of-the-art performance on benchmark datasets

Models can be deep without over-smoothing

Theoretical analysis supports the effectiveness of the approach

Abstract

We propose Graph Tree Networks (GTNets), a deep graph learning architecture with a new general message passing scheme that originates from the tree representation of graphs. In the tree representation, messages propagate upward from the leaf nodes to the root node, and each node preserves its initial information prior to receiving information from its child nodes (neighbors). We formulate a general propagation rule following the nature of message passing in the tree to update a node's feature by aggregating its initial feature and its neighbor nodes' updated features. Two graph representation learning models are proposed within this GTNet architecture - Graph Tree Attention Network (GTAN) and Graph Tree Convolution Network (GTCN), with experimentally demonstrated state-of-the-art performance on several popular benchmark datasets. Unlike the vanilla Graph Attention Network (GAT) and Graph Convolution Network (GCN) which have the "over-smoothing" issue, the proposed GTAN and GTCN models can go deep as demonstrated by comprehensive experiments and rigorous theoretical analysis.