HyperFuse: Fast Self-Supervised Node Embeddings for Attributed Hypergraphs

2026-10-02

AI Summary

Existing self-supervised representation learning methods for hypergraphs suffer from high computational complexity, limiting their scalability to large-scale or dynamic scenarios. This work proposes HyperFuse, a rapid label-free learning framework that computes structural coordinates via spectral relaxation and constructs multi-scale feature summaries. By introducing matrix-free operators, the method reduces computational complexity to linear time. Furthermore, it incorporates membership stability weighting and an invariance-decorrelation objective, leveraging a lightweight encoder to efficiently generate node embeddings. Experimental results demonstrate that HyperFuse requires only 8.7 seconds on average, achieving a 13- to 179-fold speedup over baselines while attaining state-of-the-art accuracy across most classification tasks. This approach establishes a significant balance between computational efficiency and representation performance.

Abstract

Self-supervised hypergraph representation learning can produce informative node embeddings, but existing methods often require deep encoders trained for hundreds of epochs, making embedding generation costly even for hypergraphs with a few thousand nodes. This limits applications requiring embeddings for many or evolving hypergraphs. We present HyperFuse, a label-free pipeline for fast hypergraph representation learning. HyperFuse (i) computes structural node coordinates by maximizing a spectral relaxation of hypergraph modularity using Banerjee's hypergraph adjacency and a matrix-free operator with cost linear in node-hyperedge incidences; (ii) constructs multi-scale feature summaries and assigns bounded utility weights to hyperedges based on member stability under feature and membership masking; and (iii) trains a lightweight utility-weighted hypergraph encoder for 100 epochs using an invariance-decorrelation objective. We compare HyperFuse with TriCL, SE-HSSL, VilLain, and HypeBoy on nine public hypergraphs using six downstream classifiers and k-means clustering. On the eight datasets where all methods completed, HyperFuse required 8.7 s per dataset on average, achieving 13-179x geometric-mean speed-ups over the baselines. It achieved the highest average accuracy with five of six classifiers, while classification and clustering performance was not significantly different from TriCL and SE-HSSL. Compared with HypeBoy, HyperFuse was 13x faster and 2.1-4.1 percentage points more accurate across all classifiers. HyperFuse provides a practical approach for fast, repeated hypergraph embedding generation.

Problem

• Hypergraph Representation Learning
• Self-Supervised Learning
• Node Embeddings
• Computational Efficiency

Innovation

• Hypergraph Representation Learning
• Self-Supervised Learning
• Spectral Relaxation
• Utility-Weighted Encoder
• Invariance-Decorrelation Objective