Graph Classification Gaussian Processes via Hodgelet Spectral Features

TL;DR

This paper introduces a Gaussian process-based graph classification method that utilizes spectral features derived from the Hodge decomposition to incorporate both vertex and edge features, enhancing classification performance.

Contribution

It presents a novel Gaussian process approach that leverages Hodgelet spectral features to effectively incorporate vertex and edge features in graph classification.

Findings

Improved classification accuracy on diverse graph datasets.

Effective integration of vertex and edge features using spectral methods.

Demonstrated advantages over traditional graph neural networks.

Abstract

The problem of classifying graphs is ubiquitous in machine learning. While it is standard to apply graph neural networks or graph kernel methods, Gaussian processes can be employed by transforming spatial features from the graph domain into spectral features in the Euclidean domain, and using them as the input points of classical kernels. However, this approach currently only takes into account features on vertices, whereas some graph datasets also support features on edges. In this work, we present a Gaussian process-based classification algorithm that can leverage one or both vertex and edges features. Furthermore, we take advantage of the Hodge decomposition to better capture the intricate richness of vertex and edge features, which can be beneficial on diverse tasks.