第6期 王书舟，等：支持向量机的训练算法综述 ·473· 似算法[J].计算机学报，2005,28(10)：1664-1670. [26]ZENG Xiangyan,CHEN Xuewen.SMO-based pruning ZHOU Shuisheng,ZHAN Haisheng,ZHOU Lihua.A Hu- methods for sparse least squares support vector machines ber approximation method for training the support vector [J].IEEE Trans on Neural Networks,2005,16 (6): machines[J].Chinese Journal of Computers,2005,28 1541-1546. (10):1664-1670. [27]NORIKAZU T,TETSUO N.Rigorous proof of termination [15]业宁，孙瑞样，董逸生.多拉格朗日乘子协同优化的 of SMO algorithm for support vector machines[J].IEEE SVM快速学习算法研究[J].计算机研究与发展， Trans on Neural Networks,2005,16(3):774-776. 2006,43(3):442-448. [28]GUO J,TAKAHASHI N,NISHI T.A novel sequential YE Ning,SUN Ruixiang,DONG Yisheng.SVM fast train- minimal optimization algorithm for support vector regression ing algorithm research based on multi-lagrange multiplier [C]//Proceedings of the 13th International Conference on [J].Joumal of Computer Research and Development, Neural Information Processing.Hong Kong,China,2006, 2006,43(3):442-448. 4234:827-836. [16]杨晓伟，路节，张广全.一种高效的最小二乘支持向 [29]TAKAHASHI N,GUO J,NISHI T.Global convergence of 量机分类器剪枝算法[J].计算机研究与发展，2007， SMO algorithm for support vector regression[J].IEEE 44(7):1128-1136. Trans on Neural Networks,2008,19(6):971-982. YANG Xiaowei,LU Jie,ZHANG Guangquan.An effec- [30]CAO L J,KEERTHI S S,ONG C J.Developing parallel tive pruning algorithm for least squares support vector ma- sequential minimal optimization for fast training support chine classifier[J].Journal of Computer Research and De- vector machine[J].Neurocomputing,2006,70(3):93- velopment,2007,44(7):1128-1136. 104. [17]PLATT J C.Fast training of support vector machines using [31]CAO L J,KEERTHI SS,ONG C J,et al.Parallel se- sequential minimal optimization[C//Advances in Kemnel quential minimal optimization for the training of support Methods-Support Vector Learning.Cambridge,MA:MIT vector machines J].IEEE Trans on Neural Networks, Press,1999:185-208. 2006,17(4):1039-1049. [18]KEERTHI S,SHEVADE S.BHATTCHARYYA C.et al. [32]CHEN P H,FAN R E,LIN C J.A study on SMO-type Improvements to Platt's SMO algorithm for SVM classifier decomposition methods for support vector machines[J]. design[J].Neural Computation,2001,13(3):637-649. IEEE Trans on Neural Networks,2006,17(4):893-908. [19]KEERTHI S,GILBERT E.Convergence of a generalized [33 ]BO Liefeng,JIAO Licheng,WANG Ling.Working set se- SMO algorithm for SVM classifier design J].Machine lection using functional gain for LS-SVM[J].IEEE Trans Learning,2002,46(1/2/3):351-360. on Neural Networks,2007,18(5):1541-1544. [20]LIN C J.Asymptotic convergence of an SMO algorithm [34]孙剑，郑南宁，张志华.一种训练支撑向量机的改进 without any assumptions[J].IEEE Trans on Neural Net- 贯序最小优化算法[J].软件学报，2002,13(10)： work3,2002,13(1):248-250. 2007-2012. [21]SMOLA A,SCHOLKOPF B.A tutorial on support vector SUN Jian,ZHENG Nanning,ZHANG Zhihua.An im- regressions[J].Statistics and Computing,2004,14(8): proved sequential minimization optimization algorithm for 199-222. support vector machine training[J].Joumnal of Software, 22]SHEVADE S K,KEERTHI SS,BHATTACHARYYA C. 2002,13(10):2007-2012. Improvements to SMO algorithm for SVM regression[J]. [35]李建民，张钹，林福宗.序贯最小优化的改进算法 IEEE Trans on Neural Networks,2000,11 (5):1188- [J].软件学报，2003,14(5)：918-924. 1193. LI Jianmin,ZHANG Bo,LIN Fuzong.An improvement al- [23]FLAKE G W,LAWRENCE S.Efficient SVM regression gorithm to sequential minimal optimization[J].Journal of training with SMO[J].Machine Learning,2002,46(1/ Software,2003,14(5):918-924. 2/3):271-290. [36]张浩然，韩正之回归支持向量机的改进序列最小优 [24]VOGT M.SMO algorithms for support vector machines 化学习算法[J].软件学报，2003,14(12)：2006 without bias term[R].Darmstadt,Germany:Institute of 2013. Automatic Control Laboratory for Control Systems and ZHANG Haoran,HAN Zhengzhi.An improved sequential Process Automation,Technische Univ.Darmstadt,2002. minimal optimization learning algorithm for regression sup- [25]KEERTHI SS,SHEVADE S K.SMO algorithm for least port vector machine[J].Journal of Software,2003,14 squares SVM formulations J].Neural Computation, (12):2006-2013. 2003,15(2):487-507. [37]朱齐丹，张智，邢卓异.支持向量机改进序列最小优