Random Wavelet Features for Graph Kernel Machines

2026-02-17

AI Summary

This work addresses the challenge of efficiently computing graph kernels for measuring node similarity in large-scale graphs by proposing a spectral method based on random wavelet features. It introduces random features into graph kernel approximation for the first time, particularly tailored for spectrally localized kernels. The approach constructs low-dimensional node embeddings whose inner products effectively approximate the low-rank form of any specified graph kernel, achieving significantly improved approximation accuracy while maintaining theoretical guarantees. Experimental results demonstrate that the resulting embeddings outperform existing methods across multiple graph learning tasks, offering both scalability and expressive power. This provides an efficient and theoretically grounded solution for large-scale graph representation learning.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Node embeddings map graph vertices into low-dimensional Euclidean spaces while preserving structural information. They are central to tasks such as node classification, link prediction, and signal reconstruction. A key goal is to design node embeddings whose dot products capture meaningful notions of node similarity induced by the graph. Graph kernels offer a principled way to define such similarities, but their direct computation is often prohibitive for large networks. Inspired by random feature methods for kernel approximation in Euclidean spaces, we introduce randomized spectral node embeddings whose dot products estimate a low-rank approximation of any specific graph kernel. We provide theoretical and empirical results showing that our embeddings achieve more accurate kernel approximations than existing methods, particularly for spectrally localized kernels. These results demonstrate the effectiveness of randomized spectral constructions for scalable and principled graph representation learning.

Problem

• graph kernel
• node embedding
• large-scale graphs
• kernel approximation
• scalable representation learning

Innovation

• random wavelet features
• graph kernel approximation
• randomized spectral embeddings
• node embeddings
• scalable graph representation