Message Passing Graph Kernels
This work addresses graph learning tasks for researchers and practitioners, but it is incremental as it builds on existing message passing schemes.
The paper tackles the problem of graph similarity and learning by proposing a general framework for designing graph kernels based on the message passing scheme, resulting in four derived instances that are competitive with state-of-the-art methods in various tasks.
Graph kernels have recently emerged as a promising approach for tackling the graph similarity and learning tasks at the same time. In this paper, we propose a general framework for designing graph kernels. The proposed framework capitalizes on the well-known message passing scheme on graphs. The kernels derived from the framework consist of two components. The first component is a kernel between vertices, while the second component is a kernel between graphs. The main idea behind the proposed framework is that the representations of the vertices are implicitly updated using an iterative procedure. Then, these representations serve as the building blocks of a kernel that compares pairs of graphs. We derive four instances of the proposed framework, and show through extensive experiments that these instances are competitive with state-of-the-art methods in various tasks.