LOHA: Direct Graph Spectral Contrastive Learning Between Low-pass and High-pass Views

TL;DR

LOHA introduces a contrastive learning framework that leverages low-pass and high-pass spectral views in graphs, harmonizing their differences to improve node representation learning across diverse graph structures.

Contribution

LOHA proposes a novel spectral contrastive learning method that combines low-pass and high-pass views using a new spectral signal trend, enhancing spectral GNN performance.

Findings

Achieves 2.8% average improvement over baselines on 9 datasets

Outperforms some fully-supervised models on several datasets

Effectively handles graphs with varying homophily levels

Abstract

Spectral Graph Neural Networks effectively handle graphs with different homophily levels, with low-pass filter mining feature smoothness and high-pass filter capturing differences. When these distinct filters could naturally form two opposite views for self-supervised learning, the commonalities between the counterparts for the same node remain unexplored, leading to suboptimal performance. In this paper, a simple yet effective self-supervised contrastive framework, LOHA, is proposed to address this gap. LOHA optimally leverages low-pass and high-pass views by embracing "harmony in diversity". Rather than solely maximizing the difference between these distinct views, which may lead to feature separation, LOHA harmonizes the diversity by treating the propagation of graph signals from both views as a composite feature. Specifically, a novel high-dimensional feature named spectral signal trend is proposed to serve as the basis for the composite feature, which remains relatively unaffected by changing filters and focuses solely on original feature differences. LOHA achieves an average performance improvement of 2.8% over runner-up models on 9 real-world datasets with varying homophily levels. Notably, LOHA even surpasses fully-supervised models on several datasets, which underscores the potential of LOHA in advancing the efficacy of spectral GNNs for diverse graph structures.