• 中国精品科技期刊
  • CCF推荐A类中文期刊
  • 计算领域高质量科技期刊T1类
Advanced Search
Feng Chang, Liao Shizhong. Model Selection for Gaussian Kernel Support Vector Machines in Random Fourier Feature Space[J]. Journal of Computer Research and Development, 2016, 53(9): 1971-1978. DOI: 10.7544/issn1000-1239.2016.20150489
Citation: Feng Chang, Liao Shizhong. Model Selection for Gaussian Kernel Support Vector Machines in Random Fourier Feature Space[J]. Journal of Computer Research and Development, 2016, 53(9): 1971-1978. DOI: 10.7544/issn1000-1239.2016.20150489

Model Selection for Gaussian Kernel Support Vector Machines in Random Fourier Feature Space

More Information
  • Published Date: August 31, 2016
  • Model selection is very critical to support vector machines (SVMs). Standard SVMs typically suffer from cubic time complexity in data size since they solve the convex quadratic programming problems. However, it usually needs to train hundreds/thousands of SVMs for model selection, which is prohibitively time-consuming for medium-scale datasets and very difficult to scale up to large-scale problems. In this paper, by using random Fourier features to approximate Gaussian kernel, a novel and efficient approach to model selection of kernel SVMs is proposed. Firstly, the random Fourier feature mapping is used to embed the infinite-dimensional implicit feature space into an explicit random feature space. An error bound between the accurate model obtained by training kernel SVM and the approximate one returned by the linear SVM in the random feature space is derived. Then, in the random feature space, a model selection approach to kernel SVM is presented. Under the guarantee of the model error upper bound, by applying the linear SVMs in the random feature space to approximate the corresponding kernel SVMs, the approximate model selection criterion can be efficiently calculated and used to assess the relative goodness of the corresponding kernel SVMs. Finally, comparison experiments on benchmark datasets for cross validation model selection show the proposed approach can significantly improve the efficiency of model selection for kernel SVMs while guaranteeing test accuracy.
  • Related Articles

    [1]Zhang Yong, Chen Rongrong, Zhang Jing. Safe Tri-training Algorithm Based on Cross Entropy[J]. Journal of Computer Research and Development, 2021, 58(1): 60-69. DOI: 10.7544/issn1000-1239.2021.20190838
    [2]Zhou Yu, He Jianjun, Gu Hong. Fast Kernel-Based Partial Label Learning Algorithm Based on Variational Gaussian Process Model[J]. Journal of Computer Research and Development, 2017, 54(1): 63-70. DOI: 10.7544/issn1000-1239.2017.20150796
    [3]Tian Meng, Wang Wenjian. Generalized Kernel Polarization Criterion for Optimizing Gaussian Kernel[J]. Journal of Computer Research and Development, 2015, 52(8): 1722-1734. DOI: 10.7544/issn1000-1239.2015.20150110
    [4]Liu Yong, Jiang Shali, Liao Shizhong. Approximate Gaussian Kernel for Large-Scale SVM[J]. Journal of Computer Research and Development, 2014, 51(10): 2171-2177. DOI: 10.7544/issn1000-1239.2014.20130825
    [5]Zhu Kenan, Yin Baolin, Mao Yaming, Hu Yingnan. Malware Classification Approach Based on Valid Window and Naive Bayes[J]. Journal of Computer Research and Development, 2014, 51(2): 373-381.
    [6]Yuan Chunyang, Xu Junfeng, Zhu Chunge. A Trusted Recovery Model for Assurance of Integrity Policy Validity[J]. Journal of Computer Research and Development, 2014, 51(2): 360-372.
    [7]Jiang Haitao, Li Qianmu, Xu Jian, and Zhang Hong. Cross-Regional Ferry Routing Design for Multiple Messengers in Opportunistic Networks[J]. Journal of Computer Research and Development, 2012, 49(4): 700-709.
    [8]Zhou Guangtong, Yin Yilong, Guo Xinjian, and Dong Cailing. VOTCL and the Study of Its Application on Cross-Selling Problems[J]. Journal of Computer Research and Development, 2010, 47(9): 1539-1547.
    [9]Chang Qun, Wang Xiaolong, Lin Yimeng, Daniel S. Yeung, Chen Qingcai. Reducing Gaussian Kernel's Local Risks by Global Kernel and Two-Stage Model Selection Based on Genetic Algorithms[J]. Journal of Computer Research and Development, 2007, 44(3).
    [10]Hao Zhongxiao, Han Qilong. Real-Time Multiversion Concurrency Control Based on Validation Factor[J]. Journal of Computer Research and Development, 2006, 43(3): 522-527.
  • Cited by

    Periodical cited type(25)

    1. 张世文,陈双,梁伟,李仁发. 联邦学习中的攻击手段与防御机制研究综述. 计算机工程与应用. 2024(05): 1-16 .
    2. 刘炜,刘宇昭,唐琮轲,王媛媛,佘维,田钊. 基于区块链的联邦蒸馏数据共享模型研究. 计算机科学. 2024(03): 39-47 .
    3. 汤凌韬,陈左宁,张鲁飞,吴东. 联邦学习中的隐私问题研究进展. 软件学报. 2023(01): 197-229 .
    4. 金源,李成智. 智能财务背景下的财务信息安全研究. 财会通讯. 2023(07): 136-144 .
    5. 先兴平,吴涛,乔少杰,吴渝,刘宴兵. 图学习隐私与安全问题研究综述. 计算机学报. 2023(06): 1184-1212 .
    6. 李功源,刘博涵,杨雨豪,邵栋. 可信人工智能系统的质量属性与实现:三级研究. 软件学报. 2023(09): 3941-3965 .
    7. 王守欣,彭长根,刘海,谭伟杰,张弘. 基于联邦学习的PATE教师模型聚合优化方法. 计算机与数字工程. 2023(11): 2608-2614 .
    8. 崔争艳,刘晨晨,孙滨. 基于机器学习的MOOC学习者弃学预测与预警系统实现. 信息与电脑(理论版). 2022(01): 65-67 .
    9. 王坤庆,刘婧,李晨,赵语杭,吕浩然,李鹏,刘炳莹. 联邦学习安全威胁综述. 信息安全研究. 2022(03): 223-234 .
    10. 陈玉明,董建威. 基于粒计算的非线性感知机. 数据采集与处理. 2022(03): 566-575 .
    11. 田枫,冯建臣,刘芳. 改进YOLOv4的油田作业现场烟火检测. 计算机系统应用. 2022(06): 300-306 .
    12. 宁晗阳,马苗,杨波,刘士昌. 密码学智能化研究进展与分析. 计算机科学. 2022(09): 288-296 .
    13. 刘梦君,蒋新宇,石斯瑾,江南,吴笛. 人工智能教育融合安全警示:来自机器学习算法功能的原生风险分析. 江南大学学报(人文社会科学版). 2022(05): 89-101 .
    14. 黄精武. 基于差分隐私的联邦学习数据隐私安全技术. 通信技术. 2022(12): 1618-1625 .
    15. 黄志强. 基于随机化防御的云应用安全体系技术研究. 电子设计工程. 2021(02): 150-154 .
    16. 赵俊杰,王金伟. 基于SmsGAN的对抗样本修复. 郑州大学学报(工学版). 2021(01): 50-55 .
    17. 张颖君,陈恺,周赓,吕培卓,刘勇,黄亮. 神经网络水印技术研究进展. 计算机研究与发展. 2021(05): 964-976 . 本站查看
    18. 拓世英,孙浩,林子涵,陈进. 多模态图像智能目标识别对抗攻击. 国防科技. 2021(02): 8-13 .
    19. 张宇,李海良. 基于RSA的图像可识别对抗攻击方法. 网络与信息安全学报. 2021(05): 40-48 .
    20. 黄静琪,贾西平,陈道鑫,柏柯嘉,廖秀秀. 基于双对抗机制的图像攻击算法. 计算机工程. 2021(11): 150-157 .
    21. 孙爽,李晓会,刘妍,张兴. 不同场景的联邦学习安全与隐私保护研究综述. 计算机应用研究. 2021(12): 3527-3534 .
    22. 周俊,方国英,吴楠. 联邦学习安全与隐私保护研究综述. 西华大学学报(自然科学版). 2020(04): 9-17 .
    23. 李德权,许月,薛生. 基于动态约束自适应方法抵御高维鞍点攻击. 计算机研究与发展. 2020(09): 2001-2008 . 本站查看
    24. 魏立斐,陈聪聪,张蕾,李梦思,陈玉娇,王勤. 机器学习的安全问题及隐私保护. 计算机研究与发展. 2020(10): 2066-2085 . 本站查看
    25. 宋雪亚,王传安. 文本信息分词处理下的智能家电离线语音识别. 自动化与仪器仪表. 2020(12): 161-164 .

    Other cited types(50)

Catalog

    Article views (1947) PDF downloads (895) Cited by(75)

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return