Membership Inference Attacks on Sequence-to-Sequence Models: Is My Data In Your Machine Translation System?

被引：41
作者：
Hisamoto, Sorami [1 ]
Post, Matt [2 ]
Duh, Kevin [2 ]
机构：
[1] Works Applicat, Tokyo, Japan
[2] Johns Hopkins Univ, Baltimore, MD USA
来源：
TRANSACTIONS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS | 2020年 / 8卷 / 08期
关键词：
Computational linguistics - Computer aided language translation - Data privacy;
D O I：
10.1162/tacl_a_00299
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
Data privacy is an important issue for "machine learning as a service'' providers. We focus on the problem of membership inference attacks: Given a data sample and black-box access to a model's API, determine whether the sample existed in the model's training data. Our contribution is an investigation of this problem in the context of sequence-tosequence models, which are important in applications such as machine translation and video captioning. We define the membership inference problem for sequence generation, provide an open dataset based on state-of-the-artmachine translationmodels, and report initial results on whether thesemodels leakprivate informationagainst several kinds of membership inference attacks.
引用
收藏
页码：49 / 63
页数：15
相关论文
共 33 条
[21]   Comprehensive Privacy Analysis of Deep Learning Passive and Active White-box Inference Attacks against Centralized and Federated Learning [J].
Nasr, Milad ;
Shokri, Reza ;
Houmansadr, Amir .
2019 IEEE SYMPOSIUM ON SECURITY AND PRIVACY (SP 2019), 2019, :739-753
[22]   Machine Learning with Membership Privacy using Adversarial Regularization [J].
Nasr, Milad ;
Shokri, Reza ;
Houmansadr, Amir .
PROCEEDINGS OF THE 2018 ACM SIGSAC CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY (CCS'18), 2018, :634-646
[23]   BLEU: a method for automatic evaluation of machine translation [J].
Papineni, K ;
Roukos, S ;
Ward, T ;
Zhu, WJ .
40TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, PROCEEDINGS OF THE CONFERENCE, 2002, :311-318
[24]  
Post M., 2018, P 3 C MACH TRANSL RE, P186, DOI DOI 10.18653/V1/W18-6319
[25]   Knock Knock, Who's There? Membership Inference on Aggregate Location Data [J].
Pyrgelis, Apostolos ;
Troncoso, Carmela ;
De Cristofaro, Emiliano .
25TH ANNUAL NETWORK AND DISTRIBUTED SYSTEM SECURITY SYMPOSIUM (NDSS 2018), 2018,
[26]  
Rahman MA, 2018, TRANS DATA PRIV, V11, P61
[27]  
Salem Ahmed, 2018, ARXIV180601246
[28]  
Sennrich R, 2016, PROCEEDINGS OF THE 54TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, VOL 1, P1715
[29]   Membership Inference Attacks Against Machine Learning Models [J].
Shokri, Reza ;
Stronati, Marco ;
Song, Congzheng ;
Shmatikov, Vitaly .
2017 IEEE SYMPOSIUM ON SECURITY AND PRIVACY (SP), 2017, :3-18
[30]   Auditing Data Provenance in Text-Generation Models [J].
Song, Congzheng ;
Shmatikov, Vitaly .
KDD'19: PROCEEDINGS OF THE 25TH ACM SIGKDD INTERNATIONAL CONFERENCCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2019, :196-206
1234