您的位置:山东大学 -> 科技期刊社 -> 《山东大学学报(工学版)》

山东大学学报(工学版) ›› 2016, Vol. 46 ›› Issue (2): 1-5.doi: 10.6040/j.issn.1672-3961.2.2015.030

• 机器学习与数据挖掘 •    下一篇

核心集径向基函数极限学习机

翟俊海1,张素芳2,胡文祥3,王熙照1   

  1. 1.河北大学河北省机器学习与计算智能重点试验室, 河北 保定 071002; 2.中国气象局气象干部培训学院河北分院, 河北 保定 071000;3. 河北大学计算机科学与技术学院, 河北 保定 071002
  • 收稿日期:2015-05-18 出版日期:2016-04-20 发布日期:2015-05-18
  • 作者简介:翟俊海(1964— ),男,河北易县人,教授,博士,主要研究方向为机器学习与数据挖掘.E-mail:mczjh@126.com
  • 基金资助:
    国家自然科学基金资助项目(71371063,61170040);河北省自然科学基金资助项目(F2013201220,F2013201110);河北省高等学校科学技术研究重点资助项目(ZD20131028)

Radial basis function extreme learning machine based on core sets

ZHAI Junhai1, ZHANG Sufang2, HU Wenxiang3, WANG Xizhao1   

  1. 1. Hebei Province Key Laboratory of Machine Learning and Computational Intelligence, Hebei University, Baoding 071002, Hebei, China;
    2. Hebei Branch of Meteorological Cadres Training Institute, China Meteorological Administration, Baoding 071000, Hebei, China;
    3. School of Computer Science and Technology, Hebei University, Baoding 071002, Hebei, China
  • Received:2015-05-18 Online:2016-04-20 Published:2015-05-18

摘要: 径向基函数极限学习机(radial basis function-extreme learning machine, RBF-ELM)中的两个参数都随机地生成,这导致RBF-ELM算法的不稳定性问题。另外,对于不同的数据集,难于确定隐含层结点的个数。针对RBF-ELM的这两个问题,提出了一种改进算法。首先用核心集方法选择重要的样例,然后用选择的样例初始化中心参数,宽度参数采用随机化方法初始化。该算法不仅可以在一定程度上解决RBF-ELM的不稳定性问题,而且可以确定隐含层结点的个数。试验结果表明:该算法优于RBF-ELM。

关键词: 核函数, 极限学习机, 核心集, 径向基函数, 随机映射

Abstract: Radial basis function-extreme learning machine(RBF-ELM)employed randomized method to initialize the centers and widths. Randomly initialization of the two parameters led to instability of RBF-ELM. Moreover, for different data sets, it was difficult to determine the number of the hidden nodes. An improved algorithm was proposed, which firstly selected important instances with core set method, and then the centers were initialized with the selected instances, the width parameters were randomly initialized. The proposed algorithm not only could solve the problem of the instability of RBF-ELM to some extent, but also could determine the number of hidden layer nodes. Experimental results showed that the proposed algorithm outperforms RBF-ELM algorithm.

Key words: radial basis function, random mapping, kernel functions, core sets, extreme learning machine

中图分类号: 

  • TP181
[1] HUANG Guangbin, ZHU Qinyu, SIEW C K. Extreme learning machine: A new learning scheme of feedforward neural networks[C] //Proceedings of International Joint Conference on Neural Networks(IJCNN2004). Budapest, Hungary: IJCNN, 2004:985-990.
[2] HUANG Guangbin, ZHU Qinyu, SIEW C K. Extreme learning machine: Theory and applications[J]. Neurocomputing, 2006, 70(1-3):489-501.
[3] HUANG Guangbin, WANG Dianhui, LAN Yuan. Extreme learning machines: a survey[J]. International Journal of Machine Learning and Cybernetics, 2011, 2(2):107-122.
[4] HUANG Guangbin, CHEN Lei, SIEW C K. Universal approximation using incremental constructive feedforward networks with random hidden nodes[J]. IEEE Transactions on Neural Networks, 2006, 17(4):879-892.
[5] KUMAR S. Neural network[M]. Beijing: Tsinghua University Press, 2006.
[6] 高小伟, 蒋晓芸. BP神经网络在入侵检测系统中的应用及优化 [J]. 山东大学学报(工学版), 2006, 36(6):107-110. GAO Xiaowei, JIANG Xiaoyun. The application of BP neural network in intrusion detection system and its optimization[J]. Journal of Shandong University(Engineering Science), 2006, 36(6):107-110.
[7] 姚福安, 庞向坤, 焦营营,等. 基于三色法和BP神经网络的回转窑温度检测[J]. 山东大学学报(工学版), 2008,38(2):61-65. YAO Fuan, PANG Xiangkun, JIAO Yingying, et al. Temperature detection of a rotary kiln based on three color measurement and the BP neural network[J]. Journal of Shandong University(Engineering Science), 2008, 38(2):61-65.
[8] VAPNIK V. The nature of statistical learning theory[M]. New York: Springer, 1995.
[9] 赵加敏,冯爱民,刘学军. 局部密度嵌入的结构单类支持向量机[J]. 山东大学学报(工学版), 2012,42(4):14-18. ZHAO Jiamin, FENG Aimin, LIU Xuejun. A new structured one-class support vector machine with local density embedding[J]. Journal of Shandong University(Engineering Science), 2012, 42(4):14-18.
[10] 施珺, 朱敏. 一种基于灰色系统和支持向量机的预测优化模型[J]. 山东大学学报(工学版), 2012, 42(5):7-11. SHI Jun, ZHU Min. An optimization model for forecasting based on grey system and support vector machine [J]. Journal of Shandong University(Engineering Science), 2012, 42(5):7-11.
[11] 王雪松, 程玉虎, 郝名林. 一种支持向量机参数选择的改进分布估计算法[J]. 山东大学学报(工学版), 2009,39(3):7-10. WANG Xuesong, CHENG Yuhu, HAO Minglin. Parameters selection of a support vector machine using an improved estimation of the distribution algorithm[J]. Journal of Shandong University(Engineering Science), 2009, 39(3):7-10.
[12] HUANG Guangbin, LI Mingbin, CHEN Lei, et al. Incremental extreme learning machine with fully complex hidden nodes[J]. Neurocomputing, 2008, 71(4-6):576-583.
[13] LIANG Nanning, HUANG Guangbin, SARATCHANDRAN P, et al. A fast and accurate online sequential learning algorithm for feedforward networks[J]. IEEE Transactions on Neural Networks, 2006, 17(6):1411-1423.
[14] MICHE Y, SORJAMA A, BAS P, et al. OP-ELM: Optimally pruned extreme learning machine[J]. IEEE Transactions on Neural Networks, 2010, 21(1):158-162.
[15] HUANG Guangbin, SIEW C K. Extreme Learning Machine: RBF Network Case[C] //Proceedings of Conference on Control, Automation, Robotics and Vision. Kunming, China: CARV, 2004:1029-1036.
[16] ZHAI Junhai, XU Hongyu, WANG Xizhao. Dynamic ensemble extreme learning machine based on sample entropy[J]. Soft Computing September, 2012, 16(9):1493-1502.
[17] BADOIU M, CLARKSON K L. Optimal core-sets for balls[J]. Computational Geometry, 2008, 40(1):14-22.
[18] KUMAR P, MITCHELL J, YILDIRIM A. Approximate minimum enclosing balls in high dimensions using core-sets[J]. ACM Journal of Experimental Algorithmics, 2003(8):1-29.
[19] TSANG I W, KWOK J T, CHEUNG P M. Core vector machines: Fast SVM training on very large data sets[J]. Journal of Machine Learning Research, 2005(6):363-392.
[20] FRANK A, ASUNCION A. UCI Machine Learning Repository[EB/OL] [2013-10-05].http://archive.ics.uci.edu/ml.
[1] 张佩瑞,杨燕,邢焕来,喻琇瑛. 基于核K-means的增量多视图聚类算法[J]. 山东大学学报(工学版), 2018, 48(3): 48-53.
[2] 曹雅,邓赵红,王士同. 基于单调约束的径向基函数神经网络模型[J]. 山东大学学报(工学版), 2018, 48(3): 127-133.
[3] 何其佳,刘振丙,徐涛,蒋淑洁. 基于LBP和极限学习机的脑部MR图像分类[J]. 山东大学学报(工学版), 2017, 47(2): 86-93.
[4] 周凯,元昌安,覃晓,郑彦,冯文铎. 基于核贝叶斯压缩感知的人脸识别[J]. 山东大学学报(工学版), 2016, 46(3): 74-78.
[5] 徐庆, 段利国, 李爱萍, 阴桂梅. 基于实体词语义相似度的中文实体关系抽取[J]. 山东大学学报(工学版), 2015, 45(6): 7-15.
[6] 刘晓勇. 一种基于树核函数的半监督关系抽取方法研究[J]. 山东大学学报(工学版), 2015, 45(2): 22-26.
[7] 浩庆波, 牟少敏, 尹传环, 昌腾腾, 崔文斌. 一种基于聚类的快速局部支持向量机算法[J]. 山东大学学报(工学版), 2015, 45(1): 13-18.
[8] 鲁淑霞,李黎敏. 加权最大夹角间隔核心集向量机的不平衡数据分类[J]. 山东大学学报(工学版), 2014, 44(3): 1-7.
[9] 安春霖1,陆慧娟1,2*,郑恩辉3,王明怡1,陆羿4. 嵌入误分类代价和拒识代价的极限学习机基因表达数据分类[J]. 山东大学学报(工学版), 2013, 43(4): 18-25.
[10] 郭慧玲,王士同*,闫晓波. 基于广义旋转不变性核函数的人脸识别[J]. 山东大学学报(工学版), 2012, 42(5): 71-79.
[11] 张思懿1,2,王士同1*. 核化空间深度间距的特征提取方法[J]. 山东大学学报(工学版), 2012, 42(3): 45-51.
[12] 尹建川1,2, 邹早建1,3, 徐锋1. 一种基于Akaike信息准则的极限学习机[J]. 山东大学学报(工学版), 2011, 41(6): 7-11.
[13] 王胜春,韩捷,李剑峰,李志农 . 基于模糊域和支持向量机的故障诊断方法[J]. 山东大学学报(工学版), 2006, 36(6): 116-120 .
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!