ASPECT: Node-Level Adaptive Spectral Fusion for Graph Contrastive Learning

TL;DR

ASPECT introduces a node-level adaptive spectral fusion method for graph contrastive learning, improving representation quality by customizing spectral views for individual nodes.

Contribution

It proposes a novel adaptive spectral fusion approach that learns node-wise spectral policies and a stability-aware extension for enhanced graph contrastive learning.

Findings

ASPECT outperforms existing spectral and graph contrastive baselines on various benchmarks.

ASPECT-S further improves performance with stability-aware perturbations.

Node-level spectral adaptation enhances representation quality.

Abstract

Spectral graph contrastive learning often constructs low- and high-frequency views to capture complementary graph signals, but these views are commonly combined by graph-level or node-agnostic fusion rules. We show that graph-level fusion can incur irreducible regret on mixed graphs with separated node-wise spectral preferences. Motivated by this result, we propose ASPECT, a spectral graph contrastive learning method that adaptively fuses low- and high-frequency views at the node level. ASPECT learns a node-wise spectral policy and regularizes it using channel-wise contrastive evidence, enabling different nodes to use different spectral mixtures. We further introduce ASPECT-S, an optional stability-aware extension that uses generated graph-structure and feature perturbations to obtain empirical channel-wise sensitivity estimates, together with a Rayleigh-based spectral search bias for producing informative perturbations. Experiments on homophilic and heterophilic benchmarks show that ASPECT improves representation quality over competitive spectral and graph contrastive baselines, while ASPECT-S further improves performance under joint graph-structure and feature perturbations.