Spectral Heterogeneous Graph Convolutions via Positive Noncommutative Polynomials

TL;DR

This paper introduces PSHGCN, a spectral graph convolution method for heterogeneous graphs that guarantees valid filter learning, backed by theoretical analysis and extensive experiments showing superior performance and scalability.

Contribution

It proposes a novel positive spectral convolution framework using noncommutative polynomials, enabling flexible and theoretically grounded heterogeneous graph filter learning.

Findings

PSHGCN outperforms baseline methods on benchmark datasets.

The method effectively learns diverse heterogeneous graph filters.

PSHGCN scales efficiently to large graphs with millions of nodes.

Abstract

Heterogeneous Graph Neural Networks (HGNNs) have gained significant popularity in various heterogeneous graph learning tasks. However, most existing HGNNs rely on spatial domain-based methods to aggregate information, i.e., manually selected meta-paths or some heuristic modules, lacking theoretical guarantees. Furthermore, these methods cannot learn arbitrary valid heterogeneous graph filters within the spectral domain, which have limited expressiveness. To tackle these issues, we present a positive spectral heterogeneous graph convolution via positive noncommutative polynomials. Then, using this convolution, we propose PSHGCN, a novel Positive Spectral Heterogeneous Graph Convolutional Network. PSHGCN offers a simple yet effective method for learning valid heterogeneous graph filters. Moreover, we demonstrate the rationale of PSHGCN in the graph optimization framework. We conducted an extensive experimental study to show that PSHGCN can learn diverse heterogeneous graph filters and outperform all baselines on open benchmarks. Notably, PSHGCN exhibits remarkable scalability, efficiently handling large real-world graphs comprising millions of nodes and edges. Our codes are available at https://github.com/ivam-he/PSHGCN.