计算机科学 ›› 2017, Vol. 44 ›› Issue (8): 242-245.doi: 10.11896/j.issn.1002-137X.2017.08.041
周闯,范彬,朱蕾,陆新江
ZHOU Chuang, FAN Bin, ZHU Lei and LU Xin-jiang
摘要: 极限学习机(ELM)在机器学习领域获得了很多的关注,并在应用方面取得了极大的成功。然而,极限学习机对训练数据中的异常值点和非高斯噪声非常敏感,从而大大阻碍了ELM的应用。概率权重ELM方法主要对含有异常值和非高斯噪声数据集进行建模,首先建立概率局部ELM模型,并在此基础上利用Parzen窗方法建立局部模型的概率分布,然后将概率分布作为权重来融合所有的局部模型以建立全局鲁棒性模型。该方法成功地应用了数学例子和UCI实例,并与传统ELM、正则化ELM和鲁棒ELM进行了比较分析,结果表明概率权重ELM表现出了较好的建模性能。
[1] CHUANG C C,LEE Z J.Hybrid robust support vector ma-chines for regression with outliers[J].Applied Soft Computing,2011,11(1):64-72. [2] WEN W,HAO Z,YANG X.Robust least squares support vector machine based on recursive outlier elimination[J].Soft Computing,2010,14(11):1241-1251. [3] YOU C C,ZHANG G G.An effective search method based on large data [J].Computer Science,2013,40(6):183-186.(in Chinese) 尤川川,张桂刚.一种基于大数据的有效搜索方法[J].计算机科学,2013,40(6):183-186. [4] ZHANG G,LI H X,GAN M.Design a Wind Speed Prediction Model Using Probabilistic Fuzzy System[J].IEEE Transactions on Industrial Informatics,2012,8(4):819-827. [5] LI H X,LIU Z.A Probabilistic Neural-Fuzzy Learning System for Stochastic Modeling[J].IEEE Transactions on Fuzzy Systems,2008,16(4):898-908. [6] LIU Z,LI H X.A probabilistic fuzzy logic system for modeling and control[J].IEEE Transactions on Fuzzy Systems,2005,13(6):848-859. [7] BAI J,YANG L H,ZHANG X Y.An anti noise SVM parameter optimization method for speech recognition[J].Journal of Central South University (Natural Science Edition),2013,44(2):604-611.(in Chinese) 白静,杨利红,张雪英.一种面向语音识别的抗噪SVM参数优化方法[J].中南大学学报(自然科学版),2013,44(2):604-611. [8] HUANG G B,ZHU Q Y,SUEW C K.Extreme learning machine:Theory and applications[J].Neurocomputing,2006,70(1-3):489-501. [9] FOSSACECA J M,MAZZUCHI T A,SARKANI S.MARK-ELM:Application of a novel Multiple Kernel Learning framework for improving the robustness of Network Intrusion Detection[J].Expert Systems with Applications,2015,42(8):4062-4080. [10] FENG N Q,QIU Y H,WANG F.A new method for improving neural network generalization ability[J].Computer Science,2006,33(2):201-204.(in Chinese) 冯乃勤,邱玉辉,王芳.一种提高神经网络泛化能力的新方法[J].计算机科学,2006,33(2):201-204. [11] ZHOU L,LAI K K,YU L.Least squares support vector machines ensemble models for credit scoring[J].Expert Systems with Applications,2010,37(37):127-133. [12] ABE D S.Support Vector Machines for Pattern Classification[J].Advances in Pattern Recognition,2010,36(5):7535-7543. [13] ZHU Q Y,QIN A K,SUGANTHAN P N,et al.Evolutionary extreme learning machine[J].Pattern Recognition,2005,38(10):1759-1763. [14] LU X J,ZHOU C,HUANG M H,et al.Regularized online sequential extreme learning machine with adaptive regulation factor for time-varying nonlinear system[J].Neurocomputing,2016,174:617-626. [15] CHEN C,YAN C,LI Y.A robust weighted least squares support vector regression based on least trimmed squares[J].Neurocomputing,2015,168:941-946. [16] TIAN S F,HUANG H K.A Simplification Algorithm to Support Vector Machines for Regression[J].Journal of Software,2002,13(6):1169-1172. [17] SUYKENS J A K,BRABANTER J D,LUKAS L,et al.Weighted least squares support vector machines:robustness and sparse approximation[J].Neurocomputing,2002,48(1-4):85-105. [18] CHEN X,YANG J,LIANG J,et al.Recursive robust leastsquares support vector regression based on maximum correntropy criterion[J].Neurocomputing,2012,97(1):63-73. [19] HUANG G B,WANG D H,LAN Y.Extreme learning ma-chines:a survey[J].International Journal of Machine Learning &Cybernetics,2011,2(2):107-122. [20] BI H,LIANG H L,WANG Y.Resampling methods and machine learning[J].Journal of Computer,2009,32(5):862-877.(in Chinese) 毕华,梁洪力,王珏.重采样方法与机器学习[J].计算机学报,2009,32(5):862-877. [21] ZHANG H,MENG X W,LIU L,et al.Improved SAR imagetarget detection based on Parzen window algorithm[J].Computer Science,2015,42(S2):151-154.(in Chinese) 张颢,孟祥伟,刘磊,等.改进的基于Parzen窗算法的SAR图像目标检测[J].计算机科学,2015,42(S2):151-154. [22] LICHMAN M.UCI Machine Learning Repository[EB/OL].(2014-02-26) [2016-07-15].http://archive.ics.uci.edu/ml/datasets.html. [23] HORATA P,CHIEWCHANWATTANA S,S UNAT K.Robust extreme learning machine[J].Neurocomputing,2013,102(2):31-44. |
No related articles found! |
|