当前位置: X-MOL 学术Int. J. Mach. Learn. & Cyber. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Causative label flip attack detection with data complexity measures
International Journal of Machine Learning and Cybernetics ( IF 3.1 ) Pub Date : 2020-07-03 , DOI: 10.1007/s13042-020-01159-7
Patrick P. K. Chan , Zhimin He , Xian Hu , Eric C. C. Tsang , Daniel S. Yeung , Wing W. Y. Ng

A causative attack which manipulates training samples to mislead learning is a common attack scenario. Current countermeasures reduce the influence of the attack to a classifier with the loss of generalization ability. Therefore, the collected samples should be analyzed carefully. Most countermeasures of current causative attack focus on data sanitization and robust classifier design. To our best knowledge, there is no work to determinate whether a given dataset is contaminated by a causative attack. In this study, we formulate a causative attack detection as a 2-class classification problem in which a sample represents a dataset quantified by data complexity measures, which describe the geometrical characteristics of data. As geometrical natures of a dataset are changed by a causative attack, we believe data complexity measures provide useful information for causative attack detection. Furthermore, a two-step secure classification model is proposed to demonstrate how the proposed causative attack detection improves the robustness of learning. Either a robust or traditional learning method is used according to the existence of causative attack. Experimental results illustrate that data complexity measures separate untainted datasets from attacked ones clearly, and confirm the promising performance of the proposed methods in terms of accuracy and robustness. The results consistently suggest that data complexity measures provide the crucial information to detect causative attack, and are useful to increase the robustness of learning.



中文翻译:

具有数据复杂性度量的因果标签翻转攻击检测

操纵训练样本以误导学习的因果攻击是一种常见的攻击情形。当前的对策通过丧失泛化能力来减小攻击对分类器的影响。因此,应仔细分析收集的样品。当前引起性攻击的大多数对策都集中在数据清理和强大的分类器设计上。据我们所知,尚无任何工作可以确定给定的数据集是否被致病性攻击所污染。在这项研究中,我们将致病性攻击检测公式化为2类分类问题,其中样本代表通过数据复杂性度量量化的数据集,描述数据的几何特征。由于数据集的几何性质因因果攻击而改变,我们认为数据复杂性度量可为引起攻击的检测提供有用的信息。此外,提出了两步安全分类模型,以证明所提出的原因攻击检测如何提高学习的鲁棒性。根据因果攻击的存在,使用健壮或传统的学习方法。实验结果表明,数据复杂性度量可以清楚地将未受污染的数据集与受攻击的数据集区分开,并在准确性和鲁棒性方面证实了所提出方法的有希望的性能。结果一致表明,数据复杂性度量可提供检测因果攻击的关键信息,并有助于提高学习的鲁棒性。提出了两步安全分类模型,以证明所提出的原因攻击检测如何提高学习的鲁棒性。根据因果攻击的存在,使用健壮或传统的学习方法。实验结果表明,数据复杂性度量可以清楚地将未受污染的数据集与受攻击的数据集区分开,并在准确性和鲁棒性方面证实了所提出方法的有希望的性能。结果一致表明,数据复杂性度量可提供检测因果攻击的关键信息,并有助于提高学习的鲁棒性。提出了两步安全分类模型,以证明所提出的原因攻击检测如何提高学习的鲁棒性。根据因果攻击的存在,使用健壮或传统的学习方法。实验结果表明,数据复杂性度量可以清楚地将未受污染的数据集与受攻击的数据集区分开,并从准确性和鲁棒性方面证实了所提出方法的有希望的性能。结果一致表明,数据复杂性度量可提供检测因果攻击的关键信息,并有助于提高学习的鲁棒性。实验结果表明,数据复杂性度量可以清楚地将未受污染的数据集与受攻击的数据集区分开,并在准确性和鲁棒性方面证实了所提出方法的有希望的性能。结果一致表明,数据复杂性度量可提供检测因果攻击的关键信息,并有助于提高学习的鲁棒性。实验结果表明,数据复杂性度量可以清楚地将未受污染的数据集与受攻击的数据集区分开,并在准确性和鲁棒性方面证实了所提出方法的有希望的性能。结果一致表明,数据复杂性度量可提供检测因果攻击的关键信息,并有助于提高学习的鲁棒性。

更新日期:2020-07-05
down
wechat
bug