RetGK: Graph Kernels based on Return Probabilities of Random Walks
By Zhen Zhang, Mianzhi Wang, Yijian Xiang, Yan Huang, Arye Nehorai. 2018
120 citations recorded by the source corpus.
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 other state-of-the-art approaches in both accuracy and computational efficiency.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Information Systems
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Christopher Morris
- Jure Leskovec
- Rex Ying
- Jiaxuan You
- William L. Hamilton
- Nils M. Kriege
- Fredrik D. Johansson
- Jonathan M. Gomes-Selman
In Computer Vision and Pattern Recognition
- Christopher Morris
- Fredrik D. Johansson
- Nils M. Kriege
- Jiaxuan You
- Jure Leskovec
- Rex Ying
- William L. Hamilton
- Petra Mutzel
In Information Systems
- Jure Leskovec
- Christopher Morris
- Rex Ying
- Jiaxuan You
- Fredrik D. Johansson
- Nils M. Kriege
- William L. Hamilton
- Marinka Zitnik
Citations in the corpus
- Identity-aware Graph Neural Networks (2021, background): “Similar cycle count information is also shown to be useful in the context of graph kernels (Zhang et al. 2018).”
- A survey on graph kernels (2019, background): “The survey by Zhang et al. (2018a) focuses on kernels for graphs without attributes which is a small subset of the scope of this survey.”