Mitigating membership inference attacks via weighted smoothing
Recent advancements in deep learning have spotlighted a crucial privacy vulnerability to membership inference attack (MIA), where adversaries can determine if specific data was present in a training set, thus potentially revealing sensitive information. In this paper, we introduce a technique, weigh...
Saved in:
Main Authors: | TAN, Minghan, XIE, Xiaofei, SUN, Jun, WANG, Tianhao |
---|---|
格式: | text |
語言: | English |
出版: |
Institutional Knowledge at Singapore Management University
2023
|
主題: | |
在線閱讀: | https://ink.library.smu.edu.sg/sis_research/8613 https://ink.library.smu.edu.sg/context/sis_research/article/9616/viewcontent/MitigatingMembership_pvoa_cc_by.pdf |
標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
機構: | Singapore Management University |
語言: | English |
相似書籍
-
Differential privacy and membership inference attacks
由: Ong, Ting Yu
出版: (2023) -
Systematic classification of attackers via bounded model checking
由: ROTHSTEIN-MORRIS, Eric, et al.
出版: (2020) -
Membership Inference Attacks Against Machine Learning Models
由: Shokri, Reza, et al.
出版: (2020) -
Verifying neural networks against backdoor attacks
由: PHAM, Long Hong, et al.
出版: (2022) -
Membership Inference Attacks Against Adversarially Robust Deep Learning Models
由: Liwei Song, et al.
出版: (2020)