RetGK: Graph Kernels based on Return Probabilities of Random Walks
This work addresses a fundamental problem in graph-structured data analysis for applications like computer vision and bioinformatics, but it appears incremental as it builds on existing graph kernel methods.
The authors tackled the problem of quantifying similarities among graphs by developing a framework for computing graph kernels based on return probabilities of random walks, resulting in significant outperformance of existing state-of-the-art approaches in both accuracy and computational efficiency.
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.