ImGCL: Revisiting Graph Contrastive Learning on Imbalanced Node Classification

被引:0
|
作者
Zeng, Liang [1 ]
Li, Lanqing [2 ]
Gao, Ziqi [3 ]
Zhao, Peilin [2 ]
Li, Jian [1 ]
机构
[1] Tsinghua Univ, Inst Interdisciplinary Informat Sci IIIS, Beijing, Peoples R China
[2] Tencent AI Lab, Shenzhen, Peoples R China
[3] Hong Kong Univ Sci & Technol, Hong Kong, Peoples R China
基金
中国国家自然科学基金;
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Graph contrastive learning (GCL) has attracted a surge of attention due to its superior performance for learning node/graph representations without labels. However, in practice, the underlying class distribution of unlabeled nodes for the given graph is usually imbalanced. This highly imbalanced class distribution inevitably deteriorates the quality of learned node representations in GCL. Indeed, we empirically find that most state-of-the-art GCL methods cannot obtain discriminative representations and exhibit poor performance on imbalanced node classification. Motivated by this observation, we propose a principled GCL framework on Imbalanced node classification (ImGCL), which automatically and adaptively balances the representations learned from GCL without labels. Specifically, we first introduce the online clustering based progressively balanced sampling (PBS) method with theoretical rationale, which balances the training sets based on pseudo-labels obtained from learned representations in GCL. We then develop the node centrality based PBS method to better preserve the intrinsic structure of graphs, by upweighting the important nodes of the given graph. Extensive experiments on multiple imbalanced graph datasets and imbalanced settings demonstrate the effectiveness of our proposed framework, which significantly improves the performance of the recent state-of-the-art GCL methods. Further experimental ablations and analyses show that the ImGCL framework consistently improves the representation quality of nodes in under-represented (tail) classes.
引用
收藏
页码:11138 / 11146
页数:9
相关论文
共 50 条
  • [1] Co-Modality Graph Contrastive Learning for Imbalanced Node Classification
    Qian, Yiyue
    Zhang, Chunhui
    Zhang, Yiming
    Wen, Qianlong
    Ye, Yanfang
    Zhang, Chuxu
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
  • [2] Line graph contrastive learning for node classification
    Li, Mingyuan
    Meng, Lei
    Ye, Zhonglin
    Xiao, Yuzhi
    Cao, Shujuan
    Zhao, Haixing
    JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES, 2024, 36 (04)
  • [3] Hybrid sampling-based contrastive learning for imbalanced node classification
    Cui, Caixia
    Wang, Jie
    Wei, Wei
    Liang, Jiye
    INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS, 2023, 14 (03) : 989 - 1001
  • [4] Hybrid sampling-based contrastive learning for imbalanced node classification
    Caixia Cui
    Jie Wang
    Wei Wei
    Jiye Liang
    International Journal of Machine Learning and Cybernetics, 2023, 14 : 989 - 1001
  • [5] Graph Neural Network with curriculum learning for imbalanced node classification
    Li, Xiaohe
    Fan, Zide
    Huang, Feilong
    Hu, Xuming
    Deng, Yawen
    Wang, Lei
    Zhao, Xinyu
    NEUROCOMPUTING, 2024, 574
  • [6] GCL: Contrastive learning instead of graph convolution for node classification
    Li, Shu
    Han, Lixin
    Wang, Yang
    Pu, Yonglin
    Zhu, Jun
    Li, Jingxian
    NEUROCOMPUTING, 2023, 551
  • [7] Supervised Graph Contrastive Learning for Few-Shot Node Classification
    Tan, Zhen
    Ding, Kaize
    Guo, Ruocheng
    Liu, Huan
    MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2022, PT II, 2023, 13714 : 394 - 411
  • [8] A Unique Framework of Heterogeneous Augmentation Graph Contrastive Learning for Both Node and Graph Classification
    Shao, Qi
    Chen, Duxin
    Yu, Wenwu
    IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING, 2024, 11 (06): : 5818 - 5828
  • [9] Adaptive Unified Contrastive Learning for Imbalanced Classification
    Cong, Cong
    Yang, Yixing
    Liu, Sidong
    Pagnucco, Maurice
    Di Ieva, Antonio
    Berkovsky, Shlomo
    Song, Yang
    MACHINE LEARNING IN MEDICAL IMAGING, MLMI 2022, 2022, 13583 : 348 - 357
  • [10] Graph Contrastive Learning Method with Sample Disparity Constraint and Feature Structure Graph for Node Classification
    Chen, Gangbin
    Cheng, Junwei
    Liang, Wanying
    He, Chaobo
    Tang, Yong
    KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, PT IV, KSEM 2023, 2023, 14120 : 291 - 303