计算机科学 ›› 2015, Vol. 42 ›› Issue (9): 195-198.doi: 10.11896/j.issn.1002-137X.2015.09.037
冯昌,李子达,廖士中
FENG Chang, LI Zi-da and LIAO Shi-zhong
摘要: 现有大规模支持向量机求解算法需要大量的内存资源和训练时间,通常在大集群并行环境下才能实现。提出了一种大规模支持向量机(SVM)的高效求解算法,以在个人PC机求解大规模SVM。它包括3个步骤:首先对大规模样本进行子采样来降低数据规模;然后应用随机傅里叶映射显式地构造随机特征空间,使得可在该随机特征空间中应用线性SVM来一致逼近高斯核SVM;最后给出线性SVM在多核环境下的并行实现方法以进一步提高求解效率。标准数据集的对比实验验证了该求解算法的可行性与高效性。
[1] Vapnik V.The Nature of Statistical Learning Theory[M].New York:Springer Verlag,2000 [2] 文益民,王耀南,吕宝粮,等.支持向量机处理大规模问题算法综述[J].计算机科学,2009,36(7):20-25 Wen Yi-min,Wang Yao-nan,Lv Bao-liang,et al.Survey of Applying Support Vector Machines to Handle Large-scale Problems[J].Computer Science,2009,36(7):20-25 [3] Platt J C.Fast training of support vector machines using sequential minimal optimization[M]∥Schlkopf B,Burges C,Smola A.Advances in Kernel Methods:Support Vector Learning.Cambridge:MIT Press,1999:185-208 [4] Tsang L W,Kwok J,Cheung P M.Core vector machines:Fast SVM training on very large data sets[J].Journal of Machine Learning Research,2005,6(2):363-392 [5] Chang E Y,Zhu Kai-hua,Wang Hao,et al.Parallelizing support vector machines on distributed computers[M]∥ Platt J C,Koller D,Singer Y,et al.,eds.Advances in Neural Information Processing Systems 20.Cambridge:MIT Press,2008:257-264 [6] Zhang Tong.Solving large scale linear prediction problems using stochastic gradient descent algorithms[C]∥Proceedings of the 21st International Conference on Machine Learning,2004.New York,USA,2004:919-926 [7] Fan R E,Chang K W,Hsieh C J.LIBLINEAR:A library for large linear classification[J].Journal of Machine Learning Research,2008,9(12):1871-1874 [8] Rahimi A,Recht B.Random features for large-scale kernel machines[M]∥Platt J C,Koller D,Singer Y,et al.,eds.Advances in Neural Information Processing Systems 20.Cambridge:MIT Press,2008:1177-1184 [9] Boyd S,Parikh N,Chu E,et al.Distributed optimization and statistical learning via the alternating direction method of multip-liers[J].Foundations and Trends in Machine Learning,2011,3(1):1-122 [10] Chang Y W,Hsieh C J,Chang K W,et al.Training and testing low-degree polynomial data mappings via linear SVM[J].Journal of Machine Learning Research,2010,11(4):1471-1490 [11] Zhang K,Lan L,Wang Z,et al.Scaling up kernel SVM on limi-ted resources:A low-rank linearization approach[C]∥Procee-dings of the 15th International Conference on Artificial Intelligence and Statistics,2012.Canary,Spain,2012:1425-1434 |
No related articles found! |
|