.png){kind=icon}
返回
Data sanitization against adversarial label contamination based on data complexity
.png){kind=icon}
PRE
AI
DOI：10.1007/s13042-016-0629-5.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Machine learning techniques may suffer from adversarial attack in which an attacker misleads a learning process by manipulating training samples. Data sanitization is one of countermeasures against poisoning attack. It is a data pre-processing method which filters suspect samples before learning. Recently, a number of data sanitization methods are devised for label flip attack, but their flexibility is limited due to specific assumptions. It is observed that abrupt label flip caused by attack changes complexity of classification. A data sanitization method based on data complexity, which is a measure of the difficulty of classification on a dataset, is proposed in this paper. Our method measures the data complexity of a training set after removing a sample and its nearest samples. Contaminated samples are then distinguished from untainted samples according to their data complexity values. Experimental results support the idea that data complexity can be used to identify attack samples. The proposed method achieves a better result than the current sanitization method in terms of detection accuracy for well known security application problems.
Keyword:
Adversarial learning
Poisoning attack
Data sanitization
Data complexity
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
2.7
论文数:
3.2K
被引数:
5.6K
机构
引用论文
Thermal application of composites of iron and magnesium in thermosyphon solar water heating system
.png){kind=icon}
PREAI
分享收藏
Quantification of side-channel information leaks based on data complexity measures for web browsing
PREAI
分享收藏
查看更多内容