有效的ν支持向量回归机的ν解路径算法
作者:
基金项目:

国家自然科学基金重点项目(61139002); 国家自然科学基金青年科学基金(61202137); 江苏高校优势学科建设工程资助项目; 南京信息工程大学科研启动费(20110433)


Effective ν-Path Algorithm for ν-Support Vector Regression
Author:
  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [19]
  • |
  • 相似文献 [20]
  • | | |
  • 文章评论
    摘要:

    由 Sch?lkopf 等人提出的ν支持向量回归机具有通过参数ν控制支持向量和错误向量个数的优点,然而与标准的支持向量机相比,其形式更为复杂,迄今为止仍没有有效的算法计算ν解路径.基于ν支持向量回归机的修改形式,提出了一种新的解路径算法,它能够追踪参数ν对应的所有解,并通过理论分析和实验,说明了该算法能够尽可能地避免不可行的更新路径,并在有限步内拟合出所有的ν解路径.

    Abstract:

    The ν-support vector regression (ν-SVR) proposed by Sch lkopf, et al., has the advantage of using theparameter ν to control the number of support vectors and margin errors, however, compared to ε-SVR, itsformulation is more complicated. Until now, there have been no effective methods used to compute the ν-path for it.This paper proposes a new solution path algorithm, which is designed based on a modified formulation of ν-SVRand traces the solution path with respect to the parameter ν. Through theoretical analysis and experiments, resultscan show that the algorithm can avoid the infeasible updating path, and fit the entire ν-path in finite steps.

    参考文献
    [1] Vapnik V. Statistical Learning Theory. New York: Wiley, 1998.
    [2] Keerthi S, Sindhwani V, Chapelle O. An efficient method for gradient-based adaptation of hyperparameters in SVM models. In: Sch lkopf B, Platt J, Hoffman T, eds. Advances in Neural Information Processing Systems 19. Cambridge: MIT Press, 2007.673 680.
    [3] Bousquet O, Elisseeff A. Stability and generalization. Journal of Machine Learning Research, 2002,2(Mar.):499 526. [doi: 10.1162/153244302760200704]
    [4] Anguita D, Ridella S, Rivieccio F. K-Fold generalization capabilityassessment for support vector classifiers. In: Proc. of the Int’l Joint Conf. on Neural Networks. Montreal, 2005. 855 858. http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1555964& isnumber=33090 [doi: 10.1109/IJCNN.2005.1555964]
    [5] Hastie T, Rosset S, Tibshirani R, Zhu J. The entire regularization path for the support vector machine. Journal of Machine Learning Research, 2004,5:1391 1415.
    [6] Zhu J, Rosset S, Hastie T, Tibshirani R. 1-Norm support vector machines. In: Advances in Neural Information Processing Systems.2003. http://books.nips.cc/papers/files/nips16/NIPS2003_AA07.pdf
    [7] Gunter L, Zhu J. Efficient computation and model selection for the support vector regression. Neural Computation, 2007,19(6):1633 1655. [doi:10.1162/neco.2007.19.6.1633]
    [8] Gunter L, Zhu J. Computing the solution path for the regularized support vector regression. In: Advances in Neural Information Processing Systems, Vol.18. Cambridge: MIT Press, 2005. 483 490. http://books.nips.cc/papers/files/nips18/NIPS2005_0360.pdf
    [9] Wang G, Yeung DY, Lochovsky FH. A new solution path algorithm in support vector regression. IEEE Trans. on Neural Networks,2008,19(10):1753 1767. [doi: 10.1109/TNN.2008.2002077]
    [10] Wang G, Yeung DY, Lochovsky FH. A kernel path algorithm for support vector machines. In: Proc. of 24th Int’l Conf. on Machine Learning. 2007. 951 958. http://www.machinelearning.org/proceedings/icml2007/papers/60.pdf [doi: 10.1145/1273496.1273616]
    [11] Loosli G, Gasso G, Canu S. Regularization paths for ν-SVM and ν-SVR. In: Advances in Neural Networks Int’l Symp. on Neural Networks (ISNN). 2007. 486 496. http://www.springerlink.com/content/l1j52x77q0h15333
    [12] Sch lkopf B, Smola AJ, Williamson RC, Bartlett PL. New support vector algorithms. Neural Computation, 2000,12(5):1207 1245. [doi:10.1162/089976600300015565]
    [13] Steinwart I. On the optimal parameter choice for ν-support vector machines. IEEE Trans. on Pattern Analysis and Machine Intelligence, 2003,25(10):1274 1284. [doi: 10.1109/TPAMI.2003.1233901]
    [14] Gu B, Wang JD, Yu YC, Zheng GS, Wang LN. An effective regularization path for ν-support vector classification. In: Proc. of the2009 3rd Int’l Symp. on Intelligent Information Technology Application, Vol.02. Washington, 2009. 90 93. http://ieeexplore.ieee. org/stamp/stamp.jsp?arnumber=05370506
    [15] Chang CC, Lin CJ. Training ν-support vector regression: Theory and algorithms. Neural Computation, 2002,14(8):1959 1977. [doi:10.1162/089976602760128081]
    [16] Boyd S, Vandenberghe L. Convex Optimization. Cambridge: Cambridge University Press, 2004.
    [17] Osuna E, Freund R, Girosi F. Training support vector machines: An application to face detection. In: Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition. 1997. http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=609310 [doi: 10.1109% 2fCVPR.1997.609310]
    [18] Gu B, Wang JD, Chen HY. On-Line off-line ranking support vector machine and analysis. In: Proc. of the Int’l Joint Conf. on Neural Networks (IJCNN 2008). New York: IEEE Press, 2008. 1365 1370. http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4633975 [doi: 10.1109/IJCNN.2008.4633975]
    [19] Householder AS. The Theory of Matrices in Numerical Analysis. New York: Dover Publications, Inc., 1974. [20] Shashua A, Levin A. Taxonomy of large margin principle algorithms for ordinal regression problems. In: Advances in Neural Information Processing Systems, Vol.15. MIT Press, 2002. http://www.cs.huji.ac.il/~shashua/papers/k-planes-long.pdf
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

顾彬,王建东.有效的ν支持向量回归机的ν解路径算法.软件学报,2012,23(10):2643-2654

复制
分享
文章指标
  • 点击次数:3330
  • 下载次数: 5344
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 收稿日期:2010-01-08
  • 最后修改日期:2011-11-02
  • 在线发布日期: 2012-09-30
文章二维码
您是第19938597位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京市海淀区中关村南四街4号,邮政编码:100190
电话:010-62562563 传真:010-62562533 Email:jos@iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号