消息
×
loading..
Improved Stability Bounds for Graph Convolutional Neural Networks Under Graph Perturbations
2024
会议录名称2024 IEEE INFORMATION THEORY WORKSHOP (ITW)
ISSN2475-420X
页码307-312
发表状态已发表
DOI10.1109/ITW61385.2024.10806975
摘要

Graph convolutional neural networks (GCNNs) have emerged as powerful tools for processing signals or data supported by graphs. However, their effectiveness is compromised when perturbations exist in the graph structures. While previous research has studied the stability of GCNNs against such perturbations by analyzing the behavior of the graph convolutional filters, we found a significant discrepancy exists between the theoretical stability bounds and simulation outcomes. In this paper, we propose a novel approach to characterize the stability of GCNNs more accurately. Unlike existing methods that treat graph convolutional filters in each layer of a GCNN as separate SISO systems, our approach views them as a compact MIMO system. This perspective yields an improved stability characterization that aligns more closely with empirical observations and provides insights into designing GCNNs that are more resilient to graph perturbations. Numerical experiments on synthetic data confirm our theoretical findings and experiments on a movie recommendation problem demonstrate how it helps train stable GCNNs.

会议地点Shenzhen, China
会议日期24-28 Nov. 2024
URL查看原文
语种英语
来源库IEEE
文献类型会议论文
条目标识符https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/467849
专题信息科学与技术学院
信息科学与技术学院_硕士生
信息科学与技术学院_PI研究组_赵子平组
作者单位
School of Information Science and Technology, ShanghaiTech University, Shanghai, China
第一作者单位信息科学与技术学院
第一作者的第一单位信息科学与技术学院
推荐引用方式
GB/T 7714
Jun Zhang,Ziping Zhao. Improved Stability Bounds for Graph Convolutional Neural Networks Under Graph Perturbations[C],2024:307-312.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
个性服务
查看访问统计
谷歌学术
谷歌学术中相似的文章
[Jun Zhang]的文章
[Ziping Zhao]的文章
百度学术
百度学术中相似的文章
[Jun Zhang]的文章
[Ziping Zhao]的文章
必应学术
必应学术中相似的文章
[Jun Zhang]的文章
[Ziping Zhao]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

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。