机器学习安全攻击与防御机制研究进展和未来挑战

Xin Jiao Li; Guo Wei Wu; Lin Yao; Wei Zhe Zhang; Bin Zhang

doi:10.13328/j.cnki.jos.006147

机器学习安全攻击与防御机制研究进展和未来挑战

Translated title of the contribution: Progress and Future Challenges of Security Attacks and Defense Mechanisms in Machine Learning

Xin Jiao Li
, Guo Wei Wu^*
, Lin Yao
, Wei Zhe Zhang
, Bin Zhang

^*Corresponding author for this work

Dalian University of Technology
Peng Cheng Laboratory
School of Computer Science and Technology, Harbin Institute of Technology

Research output: Contribution to journal › Article › peer-review

15 Scopus citations

Abstract

Machine learning applications span all areas of artificial intelligence, but due to storage and transmission security issues and the flaws of machine learning algorithms themselves, machine learning faces a variety of security- and privacy-oriented attacks. This survey classifies the security and privacy attacks based on the location and timing of attacks in machine learning, and analyzes the causes and attack methods of data poisoning attacks, adversary attacks, data stealing attacks, and querying attacks. Furthermore, the existing security defense mechanisms are summarized. Finally, a perspective of future work and challenges in this research area are discussed.

Translated title of the contribution	Progress and Future Challenges of Security Attacks and Defense Mechanisms in Machine Learning
Original language	Chinese (Traditional)
Pages (from-to)	406-423
Number of pages	18
Journal	Ruan Jian Xue Bao/Journal of Software
Volume	32
Issue number	2
DOIs	https://doi.org/10.13328/j.cnki.jos.006147
State	Published - Feb 2021
Externally published	Yes

Access to Document

10.13328/j.cnki.jos.006147

Cite this

@article{8a78805931c448c9a52fbaef80e2225f,

title = "机器学习安全攻击与防御机制研究进展和未来挑战",

abstract = "Machine learning applications span all areas of artificial intelligence, but due to storage and transmission security issues and the flaws of machine learning algorithms themselves, machine learning faces a variety of security- and privacy-oriented attacks. This survey classifies the security and privacy attacks based on the location and timing of attacks in machine learning, and analyzes the causes and attack methods of data poisoning attacks, adversary attacks, data stealing attacks, and querying attacks. Furthermore, the existing security defense mechanisms are summarized. Finally, a perspective of future work and challenges in this research area are discussed.",

keywords = "Attack classification, Defense mechanism, Machine learning, Security and privacy",

author = "Li, \{Xin Jiao\} and Wu, \{Guo Wei\} and Lin Yao and Zhang, \{Wei Zhe\} and Bin Zhang",

year = "2021",

month = feb,

doi = "10.13328/j.cnki.jos.006147",

language = "繁体中文",

volume = "32",

pages = "406--423",

journal = "Ruan Jian Xue Bao/Journal of Software",

issn = "1000-9825",

publisher = "Chinese Academy of Sciences",

number = "2",

}

TY - JOUR

T1 - 机器学习安全攻击与防御机制研究进展和未来挑战

AU - Li, Xin Jiao

AU - Wu, Guo Wei

AU - Yao, Lin

AU - Zhang, Wei Zhe

AU - Zhang, Bin

PY - 2021/2

Y1 - 2021/2

N2 - Machine learning applications span all areas of artificial intelligence, but due to storage and transmission security issues and the flaws of machine learning algorithms themselves, machine learning faces a variety of security- and privacy-oriented attacks. This survey classifies the security and privacy attacks based on the location and timing of attacks in machine learning, and analyzes the causes and attack methods of data poisoning attacks, adversary attacks, data stealing attacks, and querying attacks. Furthermore, the existing security defense mechanisms are summarized. Finally, a perspective of future work and challenges in this research area are discussed.

AB - Machine learning applications span all areas of artificial intelligence, but due to storage and transmission security issues and the flaws of machine learning algorithms themselves, machine learning faces a variety of security- and privacy-oriented attacks. This survey classifies the security and privacy attacks based on the location and timing of attacks in machine learning, and analyzes the causes and attack methods of data poisoning attacks, adversary attacks, data stealing attacks, and querying attacks. Furthermore, the existing security defense mechanisms are summarized. Finally, a perspective of future work and challenges in this research area are discussed.

KW - Attack classification

KW - Defense mechanism

KW - Machine learning

KW - Security and privacy

UR - https://www.scopus.com/pages/publications/85101339389

U2 - 10.13328/j.cnki.jos.006147

DO - 10.13328/j.cnki.jos.006147

M3 - 文章

AN - SCOPUS:85101339389

SN - 1000-9825

VL - 32

SP - 406

EP - 423

JO - Ruan Jian Xue Bao/Journal of Software

JF - Ruan Jian Xue Bao/Journal of Software

IS - 2

ER -

机器学习安全攻击与防御机制研究进展和未来挑战

Abstract

Access to Document

Other files and links

Fingerprint

Cite this