Non-Homophilic Graph Pre-Training and Prompt Learning

TL;DR

This paper introduces ProNoG, a pre-training and prompt learning framework tailored for non-homophilic graphs, addressing the limitations of existing methods that overlook heterophilic characteristics, and demonstrates its effectiveness through extensive experiments.

Contribution

ProNoG is the first framework to explicitly handle non-homophilic graph characteristics in pre-training and prompt learning, with a theoretical analysis and a node-specific conditional network.

Findings

ProNoG outperforms existing methods on ten public datasets.

Theoretical insights guide the choice of pre-training tasks.

Node-specific patterns improve downstream task performance.

Abstract

Graphs are ubiquitous for modeling complex relationships between objects across various fields. Graph neural networks (GNNs) have become a mainstream technique for graph-based applications, but their performance heavily relies on abundant labeled data. To reduce labeling requirement, pre-training and prompt learning has become a popular alternative. However, most existing prompt methods do not differentiate homophilic and heterophilic characteristics of real-world graphs. In particular, many real-world graphs are non-homophilic, not strictly or uniformly homophilic with mixing homophilic and heterophilic patterns, exhibiting varying non-homophilic characteristics across graphs and nodes. In this paper, we propose ProNoG, a novel pre-training and prompt learning framework for such non-homophilic graphs. First, we analyze existing graph pre-training methods, providing theoretical insights into the choice of pre-training tasks. Second, recognizing that each node exhibits unique non-homophilic characteristics, we propose a conditional network to characterize the node-specific patterns in downstream tasks. Finally, we thoroughly evaluate and analyze ProNoG through extensive experiments on ten public datasets.