Through the Dual-Prism: A Spectral Perspective on Graph Data Augmentation for Graph Classification
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915127295475712 |
|---|---|
| author | Xia, Yutong Yu, Runpeng Liang, Yuxuan Bresson, Xavier Wang, Xinchao Zimmermann, Roger |
| author_facet | Xia, Yutong Yu, Runpeng Liang, Yuxuan Bresson, Xavier Wang, Xinchao Zimmermann, Roger |
| contents | Graph Neural Networks have become the preferred tool to process graph data, with their efficacy being boosted through graph data augmentation techniques. Despite the evolution of augmentation methods, issues like graph property distortions and restricted structural changes persist. This leads to the question: Is it possible to develop more property-conserving and structure-sensitive augmentation methods? Through a spectral lens, we investigate the interplay between graph properties, their augmentation, and their spectral behavior, and observe that keeping the low-frequency eigenvalues unchanged can preserve the critical properties at a large scale when generating augmented graphs. These observations inform our introduction of the Dual-Prism (DP) augmentation methods, including DP-Noise and DP-Mask, which retain essential graph properties while diversifying augmented graphs. Extensive experiments validate the efficiency of our approach, providing a new and promising direction for graph data augmentation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_09953 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Through the Dual-Prism: A Spectral Perspective on Graph Data Augmentation for Graph Classification Xia, Yutong Yu, Runpeng Liang, Yuxuan Bresson, Xavier Wang, Xinchao Zimmermann, Roger Machine Learning Graph Neural Networks have become the preferred tool to process graph data, with their efficacy being boosted through graph data augmentation techniques. Despite the evolution of augmentation methods, issues like graph property distortions and restricted structural changes persist. This leads to the question: Is it possible to develop more property-conserving and structure-sensitive augmentation methods? Through a spectral lens, we investigate the interplay between graph properties, their augmentation, and their spectral behavior, and observe that keeping the low-frequency eigenvalues unchanged can preserve the critical properties at a large scale when generating augmented graphs. These observations inform our introduction of the Dual-Prism (DP) augmentation methods, including DP-Noise and DP-Mask, which retain essential graph properties while diversifying augmented graphs. Extensive experiments validate the efficiency of our approach, providing a new and promising direction for graph data augmentation. |
| title | Through the Dual-Prism: A Spectral Perspective on Graph Data Augmentation for Graph Classification |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2401.09953 |