Graph convolutional network with tree-guided anisotropic message passing

doi:10.1016/j.neunet.2023.06.034

	Graph convolutional network with tree-guided anisotropic message passing
	Wang, Ruixiang 1,2; Wang, Yuhu1,2 ; Zhang, Chunxia 3; Xiang, Shiming1,2 ; Pan, Chunhong1
发表期刊	NEURAL NETWORKS
ISSN	0893-6080
	2023-08-01
卷号	165 页码:909-924
通讯作者	Zhang, Chunxia(cxzhang@bit.edu.cn)
摘要	Graph Convolutional Networks (GCNs) with naive message passing mechanisms have limited performance due to the isotropic aggregation strategy. To remedy this drawback, some recent works focus on how to design anisotropic aggregation strategies with tricks on feature mapping or structure mining. However, these models still suffer from the low ability of expressiveness and long-range modeling for the needs of high performance in practice. To this end, this paper proposes a tree-guided anisotropic GCN, which applies an anisotropic aggregation strategy with competitive expressiveness and a large receptive field. Specifically, the anisotropic aggregation is decoupled into two stages. The first stage is to establish the path of the message passing on a tree-like hypergraph consisting of substructures. The second one is to aggregate the messages with constrained intensities by employing an effective gating mechanism. In addition, a novel anisotropic readout mechanism is constructed to generate representative and discriminative graph-level features for downstream tasks. Our model outperforms baseline methods and recent works on several synthetic benchmarks and datasets from different real world tasks. In addition, extensive ablation studies and theoretical analyses indicate the effectiveness of our proposed method.& COPY; 2023 Elsevier Ltd. All rights reserved.
关键词	Deep learning Graph convolutional networks Graph structure learning Anisotropic message passing
DOI	10.1016/j.neunet.2023.06.034
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program of China[2020AAA0104 903] ; National Natural Science Foundation of China[62072039] ; National Natural Science Foundation of China[62076242]
项目资助者	National Key Research and Development Program of China ; National Natural Science Foundation of China
WOS研究方向	Computer Science ; Neurosciences & Neurology
WOS类目	Computer Science, Artificial Intelligence ; Neurosciences
WOS记录号	WOS:001047915500001
出版者	PERGAMON-ELSEVIER SCIENCE LTD
引用统计
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/54129
专题	多模态人工智能系统全国重点实验室
通讯作者	Zhang, Chunxia
作者单位	1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 3.Beijing Inst Technol, Sch Comp Sci & Technol, Beijing 100081, Peoples R China
第一作者单位	模式识别国家重点实验室
推荐引用方式 GB/T 7714	Wang, Ruixiang,Wang, Yuhu,Zhang, Chunxia,et al. Graph convolutional network with tree-guided anisotropic message passing[J]. NEURAL NETWORKS,2023,165:909-924.
APA	Wang, Ruixiang,Wang, Yuhu,Zhang, Chunxia,Xiang, Shiming,&Pan, Chunhong.(2023).Graph convolutional network with tree-guided anisotropic message passing.NEURAL NETWORKS,165,909-924.
MLA	Wang, Ruixiang,et al."Graph convolutional network with tree-guided anisotropic message passing".NEURAL NETWORKS 165(2023):909-924.