TY - JOUR
T1 - Graph Contrastive Learning with Joint Spectral Augmentation of Attribute and Topology
AU - Yang, Liang
AU - Li, Zhenna
AU - Zhuo, Jiaming
AU - Liu, Jing
AU - Ma, Ziyi
AU - Wang, Chuan
AU - Wang, Zhen
AU - Cao, Xiaochun
N1 - Publisher Copyright:
© 2025, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
PY - 2025/4/11
Y1 - 2025/4/11
N2 - As an essential technique for Graph Contrastive Learning (GCL), Graph Augmentation (GA) improves the generalization capability of the GCLs by introducing different forms of the same graph. To ensure information integrity, existing GA strategies have been designed to simultaneously process the two types of information available in graphs: node attributes and graph topology. Nonetheless, these strategies tend to augment the two types of graph information separately, ignoring their correlation, resulting in limited representation ability. To overcome this drawback, this paper proposes a novel GCL framework with a Joint spectrAl augMentation, named GCL-JAM. Motivated the equivalence between the graph learning objective on an attribute graph and the spectral clustering objective on the attribute-interpolated graph, the node attributes are first abstracted as another type of node to harmonize the node attributes and graph topology. The newly constructed graph is then utilized to perform spectral augmentation to capture the correlation during augmentation. Theoretically, the proposed joint spectral augmentation is proved to perturb more inter-class edges and noise attributes compared to separate augmentation methods. Extensive experiments on homophily and heterophily graphs validate the effectiveness and universality of GCL-JAM.
AB - As an essential technique for Graph Contrastive Learning (GCL), Graph Augmentation (GA) improves the generalization capability of the GCLs by introducing different forms of the same graph. To ensure information integrity, existing GA strategies have been designed to simultaneously process the two types of information available in graphs: node attributes and graph topology. Nonetheless, these strategies tend to augment the two types of graph information separately, ignoring their correlation, resulting in limited representation ability. To overcome this drawback, this paper proposes a novel GCL framework with a Joint spectrAl augMentation, named GCL-JAM. Motivated the equivalence between the graph learning objective on an attribute graph and the spectral clustering objective on the attribute-interpolated graph, the node attributes are first abstracted as another type of node to harmonize the node attributes and graph topology. The newly constructed graph is then utilized to perform spectral augmentation to capture the correlation during augmentation. Theoretically, the proposed joint spectral augmentation is proved to perturb more inter-class edges and noise attributes compared to separate augmentation methods. Extensive experiments on homophily and heterophily graphs validate the effectiveness and universality of GCL-JAM.
UR - http://www.scopus.com/inward/record.url?scp=105003996331&partnerID=8YFLogxK
U2 - 10.1609/aaai.v39i20.35506
DO - 10.1609/aaai.v39i20.35506
M3 - 会议文章
AN - SCOPUS:105003996331
SN - 2159-5399
VL - 39
SP - 21983
EP - 21991
JO - Proceedings of the AAAI Conference on Artificial Intelligence
JF - Proceedings of the AAAI Conference on Artificial Intelligence
IS - 20
T2 - 39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025
Y2 - 25 February 2025 through 4 March 2025
ER -