What functions can Graph Neural Networks compute on random graphs? The role of Positional Encoding

被引:0
|
作者
Keriven, Nicolas [1 ]
Vaiter, Samuel [2 ]
机构
[1] CNRS, IRISA, Rennes, France
[2] CNRS, LJAD, Nice, France
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
We aim to deepen the theoretical understanding of Graph Neural Networks (GNNs) on large graphs, with a focus on their expressive power. Existing analyses relate this notion to the graph isomorphism problem, which is mostly relevant for graphs of small sizes, or studied graph classification or regression tasks, while prediction tasks on nodes are far more relevant on large graphs. Recently, several works showed that, on very general random graphs models, GNNs converge to certains functions as the number of nodes grows. In this paper, we provide a more complete and intuitive description of the function space generated by equivariant GNNs for node-tasks, through general notions of convergence that encompass several previous examples. We emphasize the role of input node features, and study the impact of node Positional Encodings (PEs), a recent line of work that has been shown to yield state-of-the-art results in practice. Through the study of several examples of PEs on large random graphs, we extend previously known universality results to significantly more general models. Our theoretical results hint at some normalization tricks, which is shown numerically to have a positive impact on GNN generalization on synthetic and real data. Our proofs contain new concentration inequalities of independent interest.
引用
收藏
页数:27
相关论文
共 50 条
  • [31] Graph Neural Network Encoding for Community Detection in Attribute Networks
    Sun, Jianyong
    Zheng, Wei
    Zhang, Qingfu
    Xu, Zongben
    IEEE TRANSACTIONS ON CYBERNETICS, 2022, 52 (08) : 7791 - 7804
  • [32] Semi-Supervised Node Classification on Graphs: Markov Random Fields vs. Graph Neural Networks
    Wang, Binghui
    Jia, Jinyuan
    Gong, Neil Zhenqiang
    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 : 10093 - 10101
  • [33] Interpretable Clustering on Dynamic Graphs with Recurrent Graph Neural Networks
    Yao, Yuhang
    Joe-Wong, Carlee
    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 : 4608 - 4616
  • [34] Generalizing Graph Neural Networks on Out-of-Distribution Graphs
    Fan, Shaohua
    Wang, Xiao
    Shi, Chuan
    Cui, Peng
    Wang, Bai
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2024, 46 (01) : 322 - 337
  • [35] GraphSMOTE: Imbalanced Node Classification on Graphs with Graph Neural Networks
    Zhao, Tianxiang
    Zhang, Xiang
    Wang, Suhang
    WSDM '21: PROCEEDINGS OF THE 14TH ACM INTERNATIONAL CONFERENCE ON WEB SEARCH AND DATA MINING, 2021, : 833 - 841
  • [36] Group link prediction in bipartite graphs with graph neural networks
    Luo, Shijie
    Li, He
    Huang, Jianbin
    Ma, Xiaoke
    Cui, Jiangtao
    Qiao, Shaojie
    Yoo, Jaesoo
    PATTERN RECOGNITION, 2025, 158
  • [37] Learning by Transference: Training Graph Neural Networks on Growing Graphs
    Cervino, Juan
    Ruiz, Luana
    Ribeiro, Alejandro
    IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2023, 71 : 233 - 247
  • [38] Learning Graph Neural Networks on Feature-Missing Graphs
    Hu, Jun
    Wang, Jinyan
    Wei, Quanmin
    Kai, Du
    Li, Xianxian
    KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, PT I, KSEM 2023, 2023, 14117 : 255 - 262
  • [39] Graph Augmentation for Neural Networks Using Matching-Graphs
    Fuchs, Mathias
    Riesen, Kaspar
    ARTIFICIAL NEURAL NETWORKS IN PATTERN RECOGNITION, ANNPR 2022, 2023, 13739 : 3 - 15
  • [40] Can Graph Neural Networks be Adequately Explained? A Survey
    Li, Xuyan
    Wang, Jie
    Yan, Zheng
    ACM COMPUTING SURVEYS, 2025, 57 (05)