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) ›› 2018, Vol. 48 ›› Issue (3): 127-133.doi: 10.6040/j.issn.1672-3961.0.2017.423

• • 上一篇    下一篇

基于单调约束的径向基函数神经网络模型

曹雅,邓赵红*,王士同   

  1. 江南大学数字媒体技术学院, 江苏 无锡 214122
  • 收稿日期:2017-05-05 出版日期:2018-06-20 发布日期:2017-05-05
  • 通讯作者: 邓赵红(1981— ),男,安徽蒙城人,教授,博导,博士,主要研究方向为人工智能与模式识别. E-mail:dengzhaohong@jiangnan.edu.cn E-mail:caoya1027@163.com
  • 作者简介:曹雅(1992— ),女,江苏盐城人,硕士研究生,主要研究方向为人工智能与模式识别. E-mail:caoya1027@163.com
  • 基金资助:
    江苏省杰出青年基金资助项目(BK20140001);国家重点研发计划资助项目(2016YFB0800803);国家自然科学基金资助项目(61772239)

An radial basis function neural network model based on monotonic constraints

CAO Ya, DENG Zhaohong*, WANG Shitong   

  1. School of Digital Media, Jiangnan University, Wuxi 214122, Jiangsu, China
  • Received:2017-05-05 Online:2018-06-20 Published:2017-05-05

PDF (PC)

631

摘要： 径向基函数(radial basis function, RBF)神经网络是一种高效的前馈式神经网络。它结构简单,具有良好的泛化能力,已经被广泛的应用于数据分类中。但是对于一些特殊的分类场景,如单调数据场景,神经网络还未充分发挥其潜能。针对此,提出单调径向基函数神经网络(monotonic radial basis function neural network, MC-RBF)。MC-RBF引入Tikhonov 正则化方法确保优化问题解的唯一性与有界性。试验结果表明,在处理具有单调性的数据集时,MC-RBF比原始的RBF神经网络具有更好的分类性能。

关键词: 径向基函数神经网络, 数据分类, Tikhonov 正则化, 单调约束, 分类性能

Abstract: Radial basis function(RBF)neural network was a type of efficient feedforward neural network, which had simple structure and good generalization ability. It had been widely used in data classification. However, for some special classification scenarios, such as the scenarios of dealing with the monotonic data, RBF neural network could not fully realize its potential. For this challenge, monotonic radial basis function neural network(MC-RBF)was proposed. The model added a prior knowledge about monotonicity which was expressed in terms of inequality based on partial order of training data. The Tikhonov regularization was introduced to MC-RBF to ensure the uniqueness and boundedness of the solution of the optimization problem. The experimental results showed that MC-RBF had better classification performance than the classical RBF neural network when dealing with monotonic datasets.

Key words: monotonic constraint, data classification, Tikhonov regularization, radial basis function neural network, classification performance

中图分类号: 

  • TP391
[1] 罗小波,王云安,肖春宝,等. RBF神经网络在遥感影像分类中的应用研究[J]. 遥感技术与应用, 2004, 16(2): 119-123. LUO Xiaobo, WANG Yunan, XIAO Chunbao, et al. The application of RBF neural network in remote sensing image classification[J]. Remote Sensing Technology and Application, 2004, 16(2): 119-123.
[2] 那文波,何宁,刘巍,等. 基于遗传算法优化的RBF神经网络的压力传感器故障诊断[J]. 煤矿机械, 2016(7): 180-183. NA Wenbo, HE Ning, LIU Wei, et al. Fault diagnosis of pressure sensor based on RBF neural network optimized by genetic algorithm[J]. Coal Mine Machinery, 2016(7):180-183.
[3] BEN-David A, STERLING L, PAO Y H. Learning and classification of monotonic ordinal concepts[J]. Computational Intelligence, 1989, 5(1): 45-49.
[4] BEN-DAVID A. Monotonicity maintenance in information-theoretic machine learning algorithms[J]. Machine Learning, 1995, 19(1): 29-43.
[5] LI S T, CHEN C C. A regularized monotonic fuzzy support vector machine model for data mining with prior knowledge[J]. IEEE Transactions on Fuzzy Systems, 2015, 23(5):1713-1727.
[6] MAES C M. A regularized active-set method for sparse convex quadratic programming[D]. Palo Alto, USA: Stanford University, 2010.
[7] VAUHKONEN M, VADASZ D, KARJALAINEN P A, et al. Tikhonov regularization and prior information in electrical impedance tomography[J]. IEEE Transactions on Medical Imaging, 1998, 17(2):285-293.
[8] TIKHONOV A N, ARSENIN V Y. Solutions of Ill-posed problems[J]. Mathematics of Computation, 1977, 32(144):491-491.
[9] ABU-MOSTAFA Y S. Hints[J]. Neural Computation, 1995, 7(4):639-671.
[10] ABU-MOSTAFA Y S. Learning from hints[J]. Complexity, 1994, 10: 165-178.
[11] WISMER D, CHATTERGY R. Introduction to nonlinear optimization[M]. Amsterdam, Netherland: North Holland, 1978:227-313.
[12] WANG J J, ZHANG C F, JING Y Y. Self-adaptive RBF neural network PID control in exhaust temperature of micro gas turbine[C] //International Conference on Machine Learning and Cybernetics. Kunming: IEEE, 2008:2131-2136.
[13] 黄涛. 基于RBF神经网络的非线性回归模型[EB/OL].(2011-3-29)[2017-3-18]. http://www.paper.edu.cn. HUANG Tao. An nonlinear regression model based on RBF neural network [EB/OL].(2011-3-29)[2017-3-18]. http://www.paper.edu.cn.
[14] 田俊峰, 张晶, 毕志明. 基于改进的 RBF神经网络的入侵检测研究[J]. 计算机工程与应用, 2008, 44(31): 135-138. TIAN Junfeng, ZHANG Jing, BI Zhiming. Research of intrusion detection based on improved RBF neural network[J]. Computer Engineering and Applications, 2008, 44(31): 135-138.
[15] NIROS A D, TSEKOURAS G E, TSOLAKIS D, et al. Hierarchical fuzzy clustering in conjunction with particle swarm optimization to efficiently design RBF neural networks[J]. Journal of Intelligent & Robotic Systems, 2015, 78(1): 105-125.
[16] NIROS A D, TSEKOURAS G E. A novel training algorithm for RBF neural network using a hybrid fuzzy clustering approach[J]. Fuzzy Sets and Systems, 2012, 193: 62-84.
[17] YUN Z, QUAN Z, CAIXIN S, et al. RBF neural network and ANFIS-based short-term load forecasting approach in real-time price environment[J]. IEEE Transactions on Power Systems, 2008, 23(3): 853-858.
[18] 唐启义, 冯明光. DPS数据处理系统: 实验设计、统计分析及模型优化[M]. 北京: 科学出版社, 2006. TANG Qiyi, FENG Mingguang. DPS data processing system: experimental design: Statistical Analysis and Modeling[M]. Beijing: Science Press, 2006.
[19] LI Y, QIANG S, ZHUANG X, et al. Robust and adaptive backstepping control for nonlinear systems using RBF neural networks[J]. IEEE Transactions on Neural Networks, 2004, 15(3): 693-701.
[20] LIN C L, WANG J F, CHEN C Y, et al. Improving the generalization performance of RBF neural networks using a linear regression technique[J]. Expert Systems with Applications, 2009, 36(10): 12049-12053.
[21] BURGES C J C. A tutorial on support vector machines for pattern recognition[J]. Data Mining & Knowledge Discovery, 1998, 2(2):121-167.
[1] 熊冰妍, 王国胤, 邓维斌. 分级式代价敏感决策树及其在手机换机预测中的应用[J]. 山东大学学报(工学版), 2015, 45(5): 36-42.
[2] 王晓初, 王士同, 包芳. 基于概率密度分布一致约束的最小最大概率机图像分类算法[J]. 山东大学学报(工学版), 2015, 45(5): 13-21.
[3] 张东波,寇涛,许海霞. 基于LDB描述子和局部空间结构匹配的快速场景辨识[J]. 山东大学学报(工学版), 2018, 48(5): 16-23.
[4] 陈海永,余力,刘辉,杨佳博,胡启迪. 基于经验小波的太阳能电池缺陷图像融合[J]. 山东大学学报(工学版), 2018, 48(5): 24-31.
[5] 牟廉明. 自适应特征选择加权k子凸包分类[J]. 山东大学学报(工学版), 2018, 48(5): 32-37.
[6] 沈冬冬,周风余,栗梦媛,王淑倩,郭仁和. 基于集成深度神经网络的室内无线定位[J]. 山东大学学报(工学版), 2018, 48(5): 95-102.
[7] 张璞,刘畅,王永. 基于特征融合和集成学习的建议语句分类模型[J]. 山东大学学报(工学版), 2018, 48(5): 47-54.
[8] 王国新,陈凤东,刘国栋. 基于彩色伪随机编码结构光特征提取方法[J]. 山东大学学报(工学版), 2018, 48(5): 55-60.
[9] 胡建平,李鑫,谢琪,李玲,张道畅. 基于Delaunay三角化的二维无约束优化EMD方法[J]. 山东大学学报(工学版), 2018, 48(5): 9-15.
[10] 李广丽,刘斌,朱涛,殷依,张红斌. 基于优选典型相关分量的跨媒体检索模型[J]. 山东大学学报(工学版), 2018, 48(5): 38-46.
[11] 吴晨谋,方志军,黄正能. 基于单目摄像头的主动式驾驶行为分析算法[J]. 山东大学学报(工学版), 2018, 48(5): 69-76.
[12] 张宪红,张春蕊. 基于六维前馈神经网络模型的图像增强算法[J]. 山东大学学报(工学版), 2018, 48(4): 10-19.
[13] 江珊珊,杨静,范丽亚. 基于PDEs的图像特征提取方法[J]. 山东大学学报(工学版), 2018, 48(4): 27-36.
[14] 窦婷婷,姚元玺,陈鹏,芦灯. 基于ATP-EMTP的电弧建模及工程仿真[J]. 山东大学学报(工学版), 2018, 48(4): 102-108.
[15] 黄劲潮. 基于快速区域建议网络的图像多目标分割算法[J]. 山东大学学报(工学版), 2018, 48(4): 20-26.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2018 《山东大学学报(工学版)》编辑部　
地址: 济南市山大南路27号（250100）　 电话: 0531-88366735　 E-mail: xbgxb@sdu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发