Saved in:
Bibliographic Details
Main Authors: Wei, Xiaofan, Zhang, Binyan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.18148
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916707550887936
author Wei, Xiaofan
Zhang, Binyan
author_facet Wei, Xiaofan
Zhang, Binyan
contents Graph contrastive learning (GCL) has garnered significant attention recently since it learns complex structural information from graphs through self-supervised learning manner. However, prevalent GCL models may suffer from performance degradation due to inappropriate contrastive signals. Concretely, they commonly generate augmented views based on random perturbation, which leads to biased essential structures due to the introduction of noise. In addition, they assign equal weight to both hard and easy sample pairs, thereby ignoring the difference in importance of the sample pairs. To address these issues, this study proposes a novel Contrastive Signal Generative Framework for Accurate Graph Learning (CSG2L) with the following two-fold ideas: a) building a singular value decomposition (SVD)-directed augmented module (SVD-aug) to obtain the global interactions as well as avoiding the random noise perturbation; b) designing a local-global dependency learning module (LGDL) with an adaptive reweighting strategy which can differentiate the effects of hard and easy sample pairs. Extensive experiments on benchmark datasets demonstrate that the proposed CSG2L outperforms the state-of-art baselines. Moreover, CSG2L is compatible with a variety of GNNs.
format Preprint
id arxiv_https___arxiv_org_abs_2504_18148
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Generative Graph Contrastive Learning Model with Global Signal
Wei, Xiaofan
Zhang, Binyan
Machine Learning
Graph contrastive learning (GCL) has garnered significant attention recently since it learns complex structural information from graphs through self-supervised learning manner. However, prevalent GCL models may suffer from performance degradation due to inappropriate contrastive signals. Concretely, they commonly generate augmented views based on random perturbation, which leads to biased essential structures due to the introduction of noise. In addition, they assign equal weight to both hard and easy sample pairs, thereby ignoring the difference in importance of the sample pairs. To address these issues, this study proposes a novel Contrastive Signal Generative Framework for Accurate Graph Learning (CSG2L) with the following two-fold ideas: a) building a singular value decomposition (SVD)-directed augmented module (SVD-aug) to obtain the global interactions as well as avoiding the random noise perturbation; b) designing a local-global dependency learning module (LGDL) with an adaptive reweighting strategy which can differentiate the effects of hard and easy sample pairs. Extensive experiments on benchmark datasets demonstrate that the proposed CSG2L outperforms the state-of-art baselines. Moreover, CSG2L is compatible with a variety of GNNs.
title A Generative Graph Contrastive Learning Model with Global Signal
topic Machine Learning
url https://arxiv.org/abs/2504.18148