A novel perturbation attack on SVM by greedy algorithm

Yaguan QIAN; Xiaohui GUAN; Shuhui WU; NBensheng YU; Dongxiao REN

doi:10.11959/j.issn.1000−0801.2019014

您当前的位置：

首页 >

文章列表页 >

A novel perturbation attack on SVM by greedy algorithm

Research and development | 更新时间：2024-06-05

- A novel perturbation attack on SVM by greedy algorithm
- Telecommunications Science Vol. 35, Issue 1, Pages: 81-89(2019)
- 作者机构：
  
  1. 浙江科技学院，浙江杭州 310023
  2. 浙江水利水电学院，浙江杭州 310018
- 作者简介：
- 基金信息：
- DOI：10.11959/j.issn.1000−0801.2019014
  CLC： TP309.2
- Published Online：2019-01，
  
  Published：20 January 2019
- 稿件说明：
移动端阅览
Yaguan QIAN, Xiaohui GUAN, Shuhui WU, et al. A novel perturbation attack on SVM by greedy algorithm[J]. Telecommunications science, 2019, 35(1): 81-89.
DOI：

Yaguan QIAN, Xiaohui GUAN, Shuhui WU, et al. A novel perturbation attack on SVM by greedy algorithm[J]. Telecommunications science, 2019, 35(1): 81-89. DOI： 10.11959/j.issn.1000−0801.2019014.

摘要

随着对机器学习安全问题的关注度不断提高，提出一种针对SVM（support vector machine，支持向量机）的攻击样本生成方法。这种攻击发生在测试阶段，通过篡改实例数据，达到欺骗SVM分类模型的目的，具有很大的隐蔽性。采用贪心策略在核空间中搜索显著性特征子集；然后将核空间中的扰动映射回输入空间，获得攻击样本。该方法通过不超过7%的小扰动量使测试样本错误地分类。对2个数据集进行实验，攻击均能取得成功。在人造数据集中，2%的扰动量下可使SVM分类器的错误率在50%以上；在MNIST数据集中，5%的扰动量可使SVM分类器错误率接近100%。

Abstract

With the increasing concern of machine learning security issues

an adversarial sample generation method for SVM was proposed. This attack occurred in the testing stage by manipulating with the sample tofool the SVM classification model. Greedy strategy was used to search for salient feature subsets in kernel space and then the perturbation in the kernel space was projected back into the input space to obtain attack samples. This method made the test samples misclassified by less than 7% perturbation. Experiments are carried out on two data sets

and both of them are successful. In the artificial data set

the classification error rate is above 50% under 2% perturbation. In MNIST data set

the classification error rate is close to 100% under 5% perturbation.

关键词

Keywords

references

SZEGEDY C , ZAREMBA W , SUTSKEVER I , et al . Intriguing properties of neural networks [Z ] . 2014 .

PAPERNOT N , MCDANIEL P , JHA S , et al . The limitations of deep learning in adversarial settings [C ] // IEEE European Symposium on Security and Privacy , March 21 , 2016 , Saarbrucken, Germany . Piscataway : IEEE Press , 2016 : 372 - 387 .

PAPERNOT N , MCDANIEL P , GOODFELLOW I , et al . Practical black-box attacks against deep learning systems using adversarial examples [J ] . arXiv: 1602.02697 , 2016 .

ABDELWAHAB M , BUSSO C . Domain adversarial for acoustic emotion recognition [J ] . IEEE/ACM Transactions on Audio, Speech, and Language Processing , 2018 : 1 .

BARRENO M , NELSON B , SEARS R , et al . Can machine learning be secure? [C ] // ACM Symposium on Information, Computer and Communications Security , March 21 , 2006 , Taipei, China . New York : ACM Press , 2006 : 16 - 25 .

CORTES C , VAPNIK V . Support-vector networks [J ] . Machine Learning , 1995 , 20 ( 3 ): 273 - 297 .

钱亚冠，关晓惠，吴淑慧，等 . 集成方法及其在流量分类中的应用s [J ] . 电信科学， 2018 , 34 ( 4 ): 41 - 48 .

QIAN Y G , GUAN X H , WU S H , et al . An approach of Bagging ensemble based on feature set and application for traffic classification [J ] . Telecomunications Science , 2018 , 34 ( 4 ): 41 - 48 .

李昊奇，应娜，郭春生，等 . 基于深度信念网络和线性单分类SVM的高维异常检测 [J ] . 电信科学， 2018 , 34 ( 1 ): 34 - 42 .

LI H Q , YING N , GUO C S , et al . High-dimensional outlier detection based on deep belief network and linear one-class SVM [J ] . Telecomunications Science , 2018 , 34 ( 1 ): 34 - 42 .

CLIFTON C W . Opportunities for private and secure machine learning [C ] // ACM Workshop on Workshop on Aisec , Oct 27 , 2008 , Alexandria, USA . New York : ACM Press , 2008 : 31 - 32 .

RIAZI M S , WEINERT C , TKACHENKO O , et al . Chameleon:a hybrid secure computation framework for machine learning applications [J ] . arXiv:1801.03239 , 2018 .

KIM M , SONG Y , WANG S , et al . Secure logistic regression based on homomorphic encryption: design and evaluation [J ] . Jmir Medical Informatics , 2018 , 6 ( 2 ).

BIGGIO B , NELSON B , LASKOV P . Support vector machines under adversarial label noise [J ] . Journal of Machine Learning Research , 2011 , 20 ( 3 ): 97 - 112 .

BIGGIO B , NELSON B , LASKOV P . Poisoning attacks against support vector machines [C ] // The 29 th International Conference on Machine Learning , June 26 - July 1 , 2012 , Edinburgh, Scotland . New York : ACM Press , 2012 : 1467 - 1474 .

LIN X , CHAN P P K . Causative attack to incremental support vector machine [C ] // International Conference on Machine Learning and Cybernetics , July 13 , 2014 , Lanzhou, China . Piscataway : IEEE Press , 2014 : 137 - 142 .

LIU W G , HU Y G . Approach for attack signatures generating based on SVM and sequencealignment [J ] . Journal of Central South University , 2012 , 43 ( 11 ): 4328 - 4332 .

CHEN Z , TONDI B , LI X , et al . A gradient-based pixel-domain attack against SVM detection of global image manipulations [C ] // Information Forensics and Security , Nov 10 , 2018 , Hong Kong, China . Piscataway : IEEE Press , 2018 : 1 - 6 .

AMRAEE S , VAFAEI A , JAMSHIDI K , et al . Abnormal event detection in crowded scenes using one-class SVM [J ] . Signal Image & Video Processing , 2018 ( 6 ): 1 - 9 .

BENMAHAMED Y , TEGUAR M , BOUBAKEUR A . Application of SVM and KNN to Duval Pentagon 1 for transformer oil diagnosis [J ] . IEEE Transactions on Dielectrics & Electrical Insulation , 2018 , 24 ( 6 ): 3443 - 3451 .

SCHNALL A , HECKMANN M . Feature-space SVM adaptation for speaker adapted word prominence detection [J ] . Computer Speech & Language , 2018 .

ZHAO R , MAO K . Semi-random projection for dimensionality reduction and extreme learning machine in high-dimensional space [J ] . IEEE Computational Intelligence Magazine , 2015 , 10 ( 3 ): 30 - 41 .

DEVORE R A , TEMLYAKOV V N . Some remarks on greedy algorithms [J ] . Advances in Computational Mathematics , 1996 , 5 ( 1 ): 173 - 187 .

LECUN Y , CORTES C . The MNIST database of handwritten digits [Z ] . 2010 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A Scheme of QoS Evaluation Based on Support Vector Machine

Research on telecom industry customer churn prediction based on explainable machine learning models

A single flow detection enabled method for DDoS attacks in IoT based on traffic feature reconstruction and mapping

Research and application of adaptive algorithm for 5G voice quality evaluation

A spatial diversity cooperative spectrum sensing system based on blockchain and deep learning

Related Author

Ruitong Cao

Xu Cao

WANG Shengjie

ZHANG Qinghong

Lixia XIE

Bingdi YUAN

Hongyu YANG

Ze HU

Related Institution

College of Statistics and Data Science, Xinjiang University of Finance and Economics

School of Safety Science and Engineering, Civil Aviation University of China

School of Information, The University of Arizona

School of Information Engineering, Yangzhou University

School of Computer Science and Technology, Civil Aviation University of China

⁰