Anchor-guided Hypergraph Condensation with Dual-level Discrimination

2026-05-11

AI Summary

Existing hypergraph compression methods suffer from structural-feature misalignment and poor trajectory efficiency due to decoupled optimization of structure and features, leading to degraded downstream performance. To address this, we propose an anchor-guided joint optimization framework that aligns structural and feature compression without repeatedly retraining hypergraph neural networks. Our approach leverages Heat Kernel PageRank for initial node representations, employs an anchor-driven hyperedge synthesis strategy, and incorporates a theoretically grounded dual-level discrimination objective. Extensive experiments demonstrate that the proposed method significantly outperforms current state-of-the-art techniques across multiple datasets, achieving substantial gains in compression efficiency while effectively preserving both downstream task performance and structural fidelity.

Abstract

The increasing prevalence of large-scale hypergraphs poses significant computational challenges for hypergraph neural network (HNN) training. To address this, hypergraph condensation (HGC) distills large real hypergraphs into compact yet informative synthetic ones, beyond graph condensation (GC) methods limited to pairwise relations. However, existing HGC methods rely on decoupled training architectures, where structure generators are pre-trained on the original hypergraph but not jointly optimized with condensed features during refinement, resulting in misaligned structures that degrade downstream utility. Moreover, trajectory-based optimization incurs substantial computational overhead in refinement, limiting condensation efficiency. To tackle these issues, we propose \textbf{A}nchor-guided \textbf{H}yper\textbf{G}raph \textbf{C}ondensation with \textbf{D}ual-level \textbf{D}iscrimination (\textbf{AHGCDD}), which consists of three key components: (1) a node initialization module based on Heat Kernel PageRank (HKPR) to encode structural knowledge into feature semantics; (2) an anchor-guided hyperedge synthesis strategy for joint optimization of condensed features and structure; (3) a theoretically grounded dual-level discrimination objective for utility-preserving condensation without redundant HNN training. Extensive experiments demonstrate the superior effectiveness and efficiency of AHGCDD.

Problem

• hypergraph condensation
• structure-feature misalignment
• computational overhead
• large-scale hypergraphs
• downstream utility degradation

Innovation

• hypergraph condensation
• anchor-guided synthesis
• dual-level discrimination
• Heat Kernel PageRank
• joint optimization