Graph Self-Supervised Learning: A Survey

被引:161
|
作者
Liu, Yixin [1 ]
Jin, Ming [1 ]
Pan, Shirui [1 ]
Zhou, Chuan [2 ]
Zheng, Yu [3 ]
Xia, Feng [4 ]
Yu, Philip S. [5 ]
机构
[1] Monash Univ, Fac IT, Dept Data Sci & AI, Clayton, Vic 3800, Australia
[2] Chinese Acad Sci, Acad Math & Syst Sci, Beijing 100045, Peoples R China
[3] La Trobe Univ, Dept Comp Sci & Informat Technol, Bundoora, Vic 3086, Australia
[4] Federat Univ, Sch Engn Informat Technol & Phys Sci, Mt Helen, Vic 3350, Australia
[5] Univ Illinois, Dept Comp Sci, Chicago, IL 60607 USA
关键词
Task analysis; Manuals; Supervised learning; Taxonomy; Natural language processing; Data models; Deep learning; Self-supervised learning; graph analytics; deep learning; graph representation learning; graph neural networks; CONVOLUTIONAL NETWORKS; NEURAL-NETWORK;
D O I
10.1109/TKDE.2022.3172903
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Deep learning on graphs has attracted significant interests recently. However, most of the works have focused on (semi-) supervised learning, resulting in shortcomings including heavy label reliance, poor generalization, and weak robustness. To address these issues, self-supervised learning (SSL), which extracts informative knowledge through well-designed pretext tasks without relying on manual labels, has become a promising and trending learning paradigm for graph data. Different from SSL on other domains like computer vision and natural language processing, SSL on graphs has an exclusive background, design ideas, and taxonomies. Under the umbrella of graph self-supervised learning, we present a timely and comprehensive review of the existing approaches which employ SSL techniques for graph data. We construct a unified framework that mathematically formalizes the paradigm of graph SSL. According to the objectives of pretext tasks, we divide these approaches into four categories: generation-based, auxiliary property-based, contrast-based, and hybrid approaches. We further describe the applications of graph SSL across various research fields and summarize the commonly used datasets, evaluation benchmark, performance comparison and open-source codes of graph SSL. Finally, we discuss the remaining challenges and potential future directions in this research field.
引用
收藏
页码:5879 / 5900
页数:22
相关论文
共 50 条
  • [1] Self-supervised Graph Learning for Recommendation
    Wu, Jiancan
    Wang, Xiang
    Feng, Fuli
    He, Xiangnan
    Chen, Liang
    Lian, Jianxun
    Xie, Xing
    [J]. SIGIR '21 - PROCEEDINGS OF THE 44TH INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL, 2021, : 726 - 735
  • [2] Graph Adversarial Self-Supervised Learning
    Yang, Longqi
    Zhang, Liangliang
    Yang, Wenjing
    [J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 34 (NEURIPS 2021), 2021, 34
  • [3] Self-supervised Graph Learning with Segmented Graph Channels
    Gao, Hang
    Li, Jiangmeng
    Zheng, Changwen
    [J]. MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2022, PT II, 2023, 13714 : 293 - 308
  • [4] Audio self-supervised learning: A survey
    Liu, Shuo
    Mallol-Ragolta, Adria
    Parada-Cabaleiro, Emilia
    Qian, Kun
    Jing, Xin
    Kathan, Alexander
    Hu, Bin
    Schuller, Bjorn W.
    [J]. PATTERNS, 2022, 3 (12):
  • [5] Self-Supervised Learning for Videos: A Survey
    Schiappa, Madeline C.
    Rawat, Yogesh S.
    Shah, Mubarak
    [J]. ACM COMPUTING SURVEYS, 2023, 55 (13S)
  • [6] A Survey on Contrastive Self-Supervised Learning
    Jaiswal, Ashish
    Babu, Ashwin Ramesh
    Zadeh, Mohammad Zaki
    Banerjee, Debapriya
    Makedon, Fillia
    [J]. TECHNOLOGIES, 2021, 9 (01)
  • [7] Graph Self-supervised Learning with Accurate Discrepancy Learning
    Kim, Dongki
    Baek, Jinheon
    Hwang, Sung Ju
    [J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
  • [8] Contrastive Self-supervised Learning for Graph Classification
    Zeng, Jiaqi
    Xie, Pengtao
    [J]. THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2021, 35 : 10824 - 10832
  • [9] Adaptive Self-Supervised Graph Representation Learning
    Gong, Yunchi
    [J]. 36TH INTERNATIONAL CONFERENCE ON INFORMATION NETWORKING (ICOIN 2022), 2022, : 254 - 259
  • [10] Self-Supervised Bidirectional Learning for Graph Matching
    Guo, Wenqi
    Zhang, Lin
    Tu, Shikui
    Xu, Lei
    [J]. THIRTY-SEVENTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL 37 NO 6, 2023, : 7784 - 7792