AH-UGC: Adaptive and Heterogeneous-Universal Graph Coarsening

2025-05-18

AI Summary

Existing graph coarsening methods suffer from two key limitations: (1) fixed-ratio coarsening necessitates repeated computations, leading to poor efficiency; and (2) they struggle to model semantic constraints among heterogeneous node and edge types. This paper introduces AdaptCoarsen—the first unified framework supporting adaptive (arbitrary-ratio, on-the-fly) and type-isolated coarsening. Its core innovation lies in a type-aware node merging strategy that jointly leverages locality-sensitive hashing and consistent hashing, ensuring both semantic consistency and structural fidelity during efficient, scalable coarsening. Extensive experiments across 23 real-world datasets—including homogeneous/heterogeneous and homophilic/heterophilic graphs—demonstrate that AdaptCoarsen significantly reduces computational overhead while comprehensively preserving both structural integrity and semantic richness of the original graph.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic 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: Bridging structured and unstructured data

Abstract

$ extbf{Graph Coarsening (GC)}$ is a prominent graph reduction technique that compresses large graphs to enable efficient learning and inference. However, existing GC methods generate only one coarsened graph per run and must recompute from scratch for each new coarsening ratio, resulting in unnecessary overhead. Moreover, most prior approaches are tailored to $ extit{homogeneous}$ graphs and fail to accommodate the semantic constraints of $ extit{heterogeneous}$ graphs, which comprise multiple node and edge types. To overcome these limitations, we introduce a novel framework that combines Locality Sensitive Hashing (LSH) with Consistent Hashing to enable $ extit{adaptive graph coarsening}$. Leveraging hashing techniques, our method is inherently fast and scalable. For heterogeneous graphs, we propose a $ extit{type isolated coarsening}$ strategy that ensures semantic consistency by restricting merges to nodes of the same type. Our approach is the first unified framework to support both adaptive and heterogeneous coarsening. Extensive evaluations on 23 real-world datasets including homophilic, heterophilic, homogeneous, and heterogeneous graphs demonstrate that our method achieves superior scalability while preserving the structural and semantic integrity of the original graph.

Problem

• Existing GC methods require recomputation for each coarsening ratio
• Prior GC approaches fail to handle heterogeneous graphs effectively
• Lack of unified framework for adaptive and heterogeneous coarsening

Innovation

• Adaptive graph coarsening via hashing techniques
• Type isolated coarsening for heterogeneous graphs
• Unified framework for diverse graph types