Data poisoning attacks against machine learning algorithms

作者：

Highlights：

• A new approach to analyze robustness of machine learning.

• Machine learning behaviors under adversarial attacks.

• Poisoning attacks on different datasets.

摘要

•A new approach to analyze robustness of machine learning.•Machine learning behaviors under adversarial attacks.•Poisoning attacks on different datasets.

论文关键词：Cybersecurity,Machine learning,Adversarial attack,Data poisoning,Label flipping attack

论文评审过程：Received 24 October 2021, Revised 8 June 2022, Accepted 5 July 2022, Available online 14 July 2022, Version of Record 18 July 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118101