RetGK: Graph Kernels based on Return Probabilities of Random Walks

TL;DR

This paper introduces RetGK, a scalable graph kernel method based on return probabilities of random walks, which effectively utilizes node attributes and outperforms existing methods in accuracy and efficiency.

Contribution

The paper presents a novel graph kernel framework leveraging return probabilities of random walks, enhancing scalability and attribute integration for graph classification.

Findings

Significantly outperforms state-of-the-art methods in accuracy.

Demonstrates high computational efficiency on large datasets.

Effectively exploits node attributes in graph similarity measurement.

Abstract

Graph-structured data arise in wide applications, such as computer vision, bioinformatics, and social networks. Quantifying similarities among graphs is a fundamental problem. In this paper, we develop a framework for computing graph kernels, based on return probabilities of random walks. The advantages of our proposed kernels are that they can effectively exploit various node attributes, while being scalable to large datasets. We conduct extensive graph classification experiments to evaluate our graph kernels. The experimental results show that our graph kernels significantly outperform existing state-of-the-art approaches in both accuracy and computational efficiency.