ASGNN: Graph Neural Networks with Adaptive Structure
2022-10-03
状态已发表
摘要

The graph neural network (GNN) models have presented impressive achievements in numerous machine learning tasks. However, many existing GNN models are shown to be vulnerable to adversarial attacks, which creates a stringent need to build robust GNN architectures. In this work, we propose a novel interpretable message passing scheme with adaptive structure (ASMP) to defend against adversarial attacks on graph structure. Layers in ASMP are derived based on optimization steps that minimize an objective function that learns the node feature and the graph structure simultaneously. ASMP is adaptive in the sense that the message passing process in different layers is able to be carried out over dynamically adjusted graphs. Such property allows more fine-grained handling of the noisy (or perturbed) graph structure and hence improves the robustness. Convergence properties of the ASMP scheme are theoretically established. Integrating ASMP with neural networks can lead to a new family of GNN models with adaptive structure (ASGNN). Extensive experiments on semi-supervised node classification tasks demonstrate that the proposed ASGNN outperforms the state-of-the-art GNN architectures in terms of classification performance under various adversarial attacks.

DOIarXiv:2210.01002
相关网址查看原文
出处Arxiv
WOS记录号PPRN:19449668
WOS类目Computer Science, Artificial Intelligence
文献类型预印本
条目标识符https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/348522
专题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
1.ShanghaiTech Univ, Shanghai, Peoples R China
2.IBM Res, New York City, NY 10003, USA
3.Fudan Univ, Shanghai, Peoples R China
推荐引用方式
GB/T 7714
Zhang, Zepeng,Lu, Songtao,Huang, Zengfeng,et al. ASGNN: Graph Neural Networks with Adaptive Structure. 2022.
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
[Zhang, Zepeng]的文章
[Lu, Songtao]的文章
[Huang, Zengfeng]的文章
百度学术
百度学术中相似的文章
[Zhang, Zepeng]的文章
[Lu, Songtao]的文章
[Huang, Zengfeng]的文章
必应学术
必应学术中相似的文章
[Zhang, Zepeng]的文章
[Lu, Songtao]的文章
[Huang, Zengfeng]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。