Learnable Hypergraph Laplacian for Hypergraph Learning

TL;DR

This paper introduces HERALD, a learnable hypergraph Laplacian adapter that enhances hypergraph neural networks by adaptively learning the hypergraph structure and capturing non-local relations, leading to improved performance.

Contribution

The paper presents HERALD, the first learnable hypergraph Laplacian adapter that dynamically optimizes hypergraph structure and incorporates self-attention for better modeling in HGCNNs.

Findings

HERALD improves node and graph classification accuracy.

The method demonstrates strong generalization across datasets.

Significant performance gains over fixed-structure hypergraph models.

Abstract

HyperGraph Convolutional Neural Networks (HGCNNs) have demonstrated their potential in modeling high-order relations preserved in graph structured data. However, most existing convolution filters are localized and determined by the pre-defined initial hypergraph topology, neglecting to explore implicit and long-ange relations in real-world data. In this paper, we propose the first learning-based method tailored for constructing adaptive hypergraph structure, termed HypERgrAph Laplacian aDaptor (HERALD), which serves as a generic plug-in-play module for improving the representational power of HGCNNs. Specifically, HERALD adaptively optimizes the adjacency relationship between hypernodes and hyperedges in an end-to-end manner and thus the task-aware hypergraph is learned. Furthermore, HERALD employs the self-attention mechanism to capture the non-local paired-nodes relation. Extensive experiments on various popular hypergraph datasets for node classification and graph classification tasks demonstrate that our approach obtains consistent and considerable performance enhancement, proving its effectiveness and generalization ability.