Growing Like a Tree: Finding Trunks From Graph Skeleton Trees
文献类型：期刊论文
| 作者 | Huang, Zhongyu1,2; Wang, Yingheng3; Li, Chaozhuo4; He, Huiguang1,2
|
| 刊名 | IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
 |
| 出版日期 | 2024-05-01 |
| 卷号 | 46期号:5页码:2838-2851 |
| 关键词 | Graph neural networks graph-level tasks long-range patterns over-squashing skeleton trees trunks |
| ISSN号 | 0162-8828 |
| DOI | 10.1109/TPAMI.2023.3336315 |
| 通讯作者 | He, Huiguang(huiguang.he@ia.ac.cn) |
| 英文摘要 | The message-passing paradigm has served as the foundation of graph neural networks (GNNs) for years, making them achieve great success in a wide range of applications. Despite its elegance, this paradigm presents several unexpected challenges for graph-level tasks, such as the long-range problem, information bottleneck, over-squashing phenomenon, and limited expressivity. In this study, we aim to overcome these major challenges and break the conventional "node- and edge-centric" mindset in graph-level tasks. To this end, we provide an in-depth theoretical analysis of the causes of the information bottleneck from the perspective of information influence. Building on the theoretical results, we offer unique insights to break this bottleneck and suggest extracting a skeleton tree from the original graph, followed by propagating information in a distinctive manner on this tree. Drawing inspiration from natural trees, we further propose to find trunks from graph skeleton trees to create powerful graph representations and develop the corresponding framework for graph-level tasks. Extensive experiments on multiple real-world datasets demonstrate the superiority of our model. Comprehensive experimental analyses further highlight its capability of capturing long-range dependencies and alleviating the over-squashing problem, thereby providing novel insights into graph-level tasks. |
| 资助项目 | National Natural Science Foundation of China |
| WOS研究方向 | Computer Science ; Engineering |
| 语种 | 英语 |
| WOS记录号 | WOS:001196751500087 |
| 出版者 | IEEE COMPUTER SOC |
| 资助机构 | National Natural Science Foundation of China |
| 源URL | [http://ir.ia.ac.cn/handle/173211/58123]  |
| 专题 | 类脑智能研究中心_神经计算及脑机交互 |
| 通讯作者 | He, Huiguang |
| 作者单位 | 1.Chinese Acad Sci, Inst Automat, Lab Brain Atlas & Brain Inspired Intelligence, State Key Lab Multimodal ArtificialIntelligence Sy, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 3.Cornell Univ, Dept Comp Sci, Ithaca, NY 14850 USA 4.Microsoft Res Asia, Dept Social Comp, Beijing 100080, Peoples R China |
| 推荐引用方式 GB/T 7714 | Huang, Zhongyu,Wang, Yingheng,Li, Chaozhuo,et al. Growing Like a Tree: Finding Trunks From Graph Skeleton Trees[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2024,46(5):2838-2851. |
| APA | Huang, Zhongyu,Wang, Yingheng,Li, Chaozhuo,&He, Huiguang.(2024).Growing Like a Tree: Finding Trunks From Graph Skeleton Trees.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,46(5),2838-2851. |
| MLA | Huang, Zhongyu,et al."Growing Like a Tree: Finding Trunks From Graph Skeleton Trees".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 46.5(2024):2838-2851. |
入库方式： OAI收割
来源：自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。