GraDE: A Graph Diffusion Estimator for Frequent Subgraph Discovery in Neural Architectures

2026-02-03

AI Summary

Existing methods for frequent subgraph mining face significant bottlenecks in efficiency and scalability: enumeration-based approaches incur prohibitive computational costs, while sampling-based techniques suffer sharp performance degradation as subgraph size increases. To address these limitations, this work proposes GraDE, a novel framework that introduces graph diffusion models to this task for the first time. GraDE learns the underlying subgraph distribution to construct a representativeness scoring mechanism and devises a diffusion-guided search algorithm that efficiently identifies large-scale frequent subgraphs while maintaining computational feasibility. Experimental results demonstrate that GraDE improves subgraph ranking accuracy by up to 114% and discovers subgraphs whose median frequency is 30 times higher than those found by state-of-the-art sampling methods.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Content-based information diffusion
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Finding frequently occurring subgraph patterns or network motifs in neural architectures is crucial for optimizing efficiency, accelerating design, and uncovering structural insights. However, as the subgraph size increases, enumeration-based methods are perfectly accurate but computationally prohibitive, while sampling-based methods are computationally tractable but suffer from a severe decline in discovery capability. To address these challenges, this paper proposes GraDE, a diffusion-guided search framework that ensures both computational feasibility and discovery capability. The key innovation is the Graph Diffusion Estimator (GraDE), which is the first to introduce graph diffusion models to identify frequent subgraphs by scoring their typicality within the learned distribution. Comprehensive experiments demonstrate that the estimator achieves superior ranking accuracy, with up to 114\% improvement compared to sampling-based baselines. Benefiting from this, the proposed framework successfully discovers large-scale frequent patterns, achieving up to 30$\times$ higher median frequency than sampling-based methods.

Problem

• frequent subgraph discovery
• neural architectures
• network motifs
• computational scalability
• subgraph enumeration

Innovation

• Graph Diffusion
• Frequent Subgraph Discovery
• Neural Architecture Analysis
• Diffusion-guided Search
• Network Motifs