Understanding Pooling in Graph Neural Networks

被引：37
|
作者：
Grattarola, Daniele [1 ]
Zambon, Daniele [1 ]
Bianchi, Filippo Maria [2 ,3 ]
Alippi, Cesare [1 ,4 ]
机构：
[1] Univ Svizzera italiana USI, Fac Informat, CH-6904 Lugano, Switzerland
[2] UiT Arctic Univ Norway, Dept Math & Stat, N-9019 Tromso, Norway
[3] Norwegian Res Ctr NORCE, N-5838 Bergen, Norway
[4] Politecn Milan, Dept Elect Informat & Bioengn, I-20133 Milan, Italy
来源：
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS | 2024年 / 35卷 / 02期
基金：
瑞士国家科学基金会;
关键词：
Task analysis; Taxonomy; Aggregates; Point cloud compression; Laplace equations; Convolution; Clustering algorithms; Dimensionality reduction; graph neural networks (GNNs);
D O I：
10.1109/TNNLS.2022.3190922
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
Many recent works in the field of graph machine learning have introduced pooling operators to reduce the size of graphs. In this article, we present an operational framework to unify this vast and diverse literature by describing pooling operators as the combination of three functions: selection, reduction, and connection (SRC). We then introduce a taxonomy of pooling operators, based on some of their key characteristics and implementation differences under the SRC framework. Finally, we propose three criteria to evaluate the performance of pooling operators and use them to investigate the behavior of different operators on a variety of tasks.
引用
收藏
页码：2708 / 2718
页数：11
相关论文
共 50 条
  • [21] A new Graph Pooling Method based on Topology and Attribute Features in Graph Neural Networks
    Xu, Mingjun
    Gao, Qi
    Pan, Feng
    Yan, Helong
    2023 35TH CHINESE CONTROL AND DECISION CONFERENCE, CCDC, 2023, : 4120 - 4125
  • [22] Training Large-Scale Graph Neural Networks via Graph Partial Pooling
    Zhang, Qi
    Sun, Yanfeng
    Wang, Shaofan
    Gao, Junbin
    Hu, Yongli
    Yin, Baocai
    IEEE TRANSACTIONS ON BIG DATA, 2025, 11 (01) : 221 - 233
  • [23] ReiPool: Reinforced Pooling Graph Neural Networks for Graph-Level Representation Learning
    Luo, Xuexiong
    Zhang, Sheng
    Wu, Jia
    Chen, Hongyang
    Peng, Hao
    Zhou, Chuan
    Li, Zhao
    Xue, Shan
    Yang, Jian
    IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2024, 36 (12) : 9109 - 9122
  • [24] Hierarchical Representation Learning in Graph Neural Networks With Node Decimation Pooling
    Bianchi, Filippo Maria
    Grattarola, Daniele
    Livi, Lorenzo
    Alippi, Cesare
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2022, 33 (05) : 2195 - 2207
  • [25] High-Order Pooling for Graph Neural Networks with Tensor Decomposition
    Hua, Chenqing
    Rabusseau, Guillaume
    Tang, Jian
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
  • [26] Physical pooling functions in graph neural networks for molecular property prediction
    Schweidtmann, Artur M.
    Rittig, Jan G.
    Weber, Jana M.
    Grohe, Martin
    Dahmen, Manuel
    Leonhard, Kai
    Mitsos, Alexander
    COMPUTERS & CHEMICAL ENGINEERING, 2023, 172
  • [27] iPool--Information-Based Pooling in Hierarchical Graph Neural Networks
    Gao, Xing
    Dai, Wenrui
    Li, Chenglin
    Xiong, Hongkai
    Frossard, Pascal
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2022, 33 (09) : 5032 - 5044
  • [28] Uniform Pooling for Graph Networks
    Qin, Jian
    Liu, Li
    Shen, Hui
    Hu, Dewen
    APPLIED SCIENCES-BASEL, 2020, 10 (18):
  • [29] Understanding Attention and Generalization in Graph Neural Networks
    Knyazev, Boris
    Taylor, Graham W.
    Amer, Mohamed R.
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 32 (NIPS 2019), 2019, 32
  • [30] CCP-GNN: Competitive Covariance Pooling for Improving Graph Neural Networks
    Zhu, Pengfei
    Li, Jialu
    Dong, Zhe
    Hu, Qinghua
    Wang, Xiao
    Wang, Qilong
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2024, : 1 - 12
12345