HNHN: Hypergraph Networks with Hyperedge Neurons

TL;DR

HNHN introduces a hypergraph convolution network with nonlinear activations and flexible normalization, enhancing classification accuracy and computational speed on real-world datasets compared to existing methods.

Contribution

The paper presents a novel hypergraph neural network framework with nonlinear activations on hyperedges and hypernodes, and a flexible normalization scheme for improved learning.

Findings

HNHN outperforms state-of-the-art methods in classification accuracy.

HNHN demonstrates faster training and inference speeds.

The normalization scheme effectively adjusts importance based on dataset characteristics.

Abstract

Hypergraphs provide a natural representation for many real world datasets. We propose a novel framework, HNHN, for hypergraph representation learning. HNHN is a hypergraph convolution network with nonlinear activation functions applied to both hypernodes and hyperedges, combined with a normalization scheme that can flexibly adjust the importance of high-cardinality hyperedges and high-degree vertices depending on the dataset. We demonstrate improved performance of HNHN in both classification accuracy and speed on real world datasets when compared to state of the art methods.