Adversarial Attack against Cross-lingual Knowledge Graph Alignment

被引:0
|
作者
Zhang, Zeru [1 ]
Zhang, Zijie [1 ]
Zhou, Yang [1 ]
Wu, Lingfei [2 ]
Wu, Sixing [3 ]
Han, Xiaoying [1 ]
Dou, Dejing [4 ,5 ]
Che, Tianshi [1 ]
Yan, Da [6 ]
机构
[1] Auburn Univ, Auburn, AL 36849 USA
[2] JD COM Silicon Valley Res Ctr, Mountain View, CA USA
[3] Peking Univ, Beijing, Peoples R China
[4] Univ Oregon, Eugene, OR 97403 USA
[5] Baidu Res, Beijing, Peoples R China
[6] Univ Alabama Birmingham, Birmingham, AL USA
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Recent literatures have shown that knowledge graph (KG) learning models are highly vulnerable to adversarial attacks. However, there is still a paucity of vulnerability analyses of cross-lingual entity alignment under adversarial attacks. This paper proposes an adversarial attack model with two novel attack techniques to perturb the KG structure and degrade the quality of deep cross-lingual entity alignment. First, an entity density maximization method is employed to hide the attacked entities in dense regions in two KGs, such that the derived perturbations are unnoticeable. Second, an attack signal amplification method is developed to reduce the gradient vanishing issues in the process of adversarial attacks for further improving the attack effectiveness.
引用
收藏
页码:5320 / 5337
页数:18
相关论文
共 50 条
  • [1] Multilingual Knowledge Graph Embeddings for Cross-lingual Knowledge Alignment
    Chen, Muhao
    Tian, Yingtao
    Yang, Mohan
    Zaniolo, Carlo
    [J]. PROCEEDINGS OF THE TWENTY-SIXTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2017, : 1511 - 1517
  • [2] Adaptive Entity Alignment for Cross-Lingual Knowledge Graph
    Zhang, Yuanming
    Gao, Tianyu
    Lu, Jiawei
    Cheng, Zhenbo
    Xiao, Gang
    [J]. KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, KSEM 2021, PT II, 2021, 12816 : 474 - 487
  • [3] Coordinated Reasoning for Cross-Lingual Knowledge Graph Alignment
    Xu, Kun
    Song, Linfeng
    Feng, Yansong
    Song, Yan
    Yu, Dong
    [J]. THIRTY-FOURTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THE THIRTY-SECOND INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE AND THE TENTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2020, 34 : 9354 - 9361
  • [4] Entity Alignment for Cross-lingual Knowledge Graph with Graph Convolutional Networks
    Xiong, Fan
    Gao, Jianliang
    [J]. PROCEEDINGS OF THE TWENTY-EIGHTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2019, : 6480 - 6481
  • [5] Cross-lingual Knowledge Graph Alignment via Graph Convolutional Networks
    Wang, Zhichun
    Lv, Qingsong
    Lan, Xiaohan
    Zhang, Yu
    [J]. 2018 CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING (EMNLP 2018), 2018, : 349 - 357
  • [6] Cross-Lingual Taxonomy Alignment with Bilingual Knowledge Graph Embeddings
    Wu, Tianxing
    Zhang, Du
    Zhang, Lei
    Qi, Guilin
    [J]. SEMANTIC TECHNOLOGY, JIST 2017, 2017, 10675 : 251 - 258
  • [7] Guiding Cross-lingual Entity Alignment via Adversarial Knowledge Embedding
    Lin, Xixun
    Yang, Hong
    Wu, Jia
    Zhou, Chuan
    Wang, Bin
    [J]. 2019 19TH IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM 2019), 2019, : 429 - 438
  • [8] Cross-lingual Knowledge Graph Alignment via Graph Matching Neural Network
    Xu, Kun
    Wang, Liwei
    Yu, Mo
    Feng, Yansong
    Song, Yan
    Wang, Zhiguo
    Yu, Dong
    [J]. 57TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2019), 2019, : 3156 - 3161
  • [9] MRAEA: An Efficient and Robust Entity Alignment Approach for Cross-lingual Knowledge Graph
    Mao, Xin
    Wang, Wenting
    Xu, Huimin
    Lan, Man
    Wu, Yuanbin
    [J]. PROCEEDINGS OF THE 13TH INTERNATIONAL CONFERENCE ON WEB SEARCH AND DATA MINING (WSDM '20), 2020, : 420 - 428
  • [10] Cross-lingual knowledge graph entity alignment by aggregating extensive structures and specific semantics
    Zhu B.
    Bao T.
    Han J.
    Han R.
    Liu L.
    Peng T.
    [J]. Journal of Ambient Intelligence and Humanized Computing, 2023, 14 (09) : 12609 - 12616