Identity-Preserving Adversarial Training for Robust Network Embedding

doi:10.1007/s11390-023-2256-4

	Identity-Preserving Adversarial Training for Robust Network Embedding
	Cen, Ke-Ting 1,2; Shen, Hua-Wei 1,2,3; Cao, Qi 1; Xu, Bing-Bing 1; Cheng, Xue-Qi 2,4
	2024-02-01
发表期刊	JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY
ISSN	1000-9000
卷号	39 期号:1 页码:177-191
摘要	Network embedding, as an approach to learning low-dimensional representations of nodes, has been proved extremely useful in many applications, e.g., node classification and link prediction. Unfortunately, existing network embedding models are vulnerable to random or adversarial perturbations, which may degrade the performance of network embedding when being applied to downstream tasks. To achieve robust network embedding, researchers introduce adversarial training to regularize the embedding learning process by training on a mixture of adversarial examples and original examples. However, existing methods generate adversarial examples heuristically, failing to guarantee the imperceptibility of generated adversarial examples, and thus limit the power of adversarial training. In this paper, we propose a novel method Identity-Preserving Adversarial Training (IPAT) for network embedding, which generates imperceptible adversarial examples with explicit identity-preserving regularization. We formalize such identity-preserving regularization as a multi-class classification problem where each node represents a class, and we encourage each adversarial example to be discriminated as the class of its original node. Extensive experimental results on real-world datasets demonstrate that our proposed IPAT method significantly improves the robustness of network embedding models and the generalization of the learned node representations on various downstream tasks.
关键词	network embedding identity-preserving adversarial training adversarial the example
DOI	10.1007/s11390-023-2256-4
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[U21B2046] ; National Natural Science Foundation of China[62102402] ; National Key Research and Development Program of China[2020AAA0105200] ; CCF-Tencent Open Research Fund[RAGR20210108] ; Beijing Academy of Artificial Intelligence (BAAI)
WOS研究方向	Computer Science
WOS类目	Computer Science, Hardware & Architecture ; Computer Science, Software Engineering
WOS记录号	WOS:001200760400007
出版者	SPRINGER SINGAPORE PTE LTD
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/38981
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Shen, Hua-Wei
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Data Intelligence Syst Res Ctr, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 101480, Peoples R China 3.Beijing Acad Artificial Intelligence, Beijing 100000, Peoples R China 4.Chinese Acad Sci, Key Lab Network Data Sci & Technol, Inst Comp Technol, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Cen, Ke-Ting,Shen, Hua-Wei,Cao, Qi,et al. Identity-Preserving Adversarial Training for Robust Network Embedding[J]. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,2024,39(1):177-191.
APA	Cen, Ke-Ting,Shen, Hua-Wei,Cao, Qi,Xu, Bing-Bing,&Cheng, Xue-Qi.(2024).Identity-Preserving Adversarial Training for Robust Network Embedding.JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY,39(1),177-191.
MLA	Cen, Ke-Ting,et al."Identity-Preserving Adversarial Training for Robust Network Embedding".JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY 39.1(2024):177-191.