A Combined Algorithm Based on ELM-RBF and Genetic Algorithm

Abstract:

Extreme Learning Machine-Radial Basis Function (ELM-RBF) not only inherit RBF’s merit of not suffering from local minima, but also ELM’s merit of fast learning speed, Nevertheless, it is still a research hot area of how to improve the generalization ability of ELM-RBF network. Genetic Algorithms (GA) to solve optimization problem has its unique advantage. Considered on these, the paper adopted GA to optimize ELM-RBF neural network hidden layer neurons center and biases value. Experiments data results indicated that our proposed combined algorithm has better generalization performance than classical ELM-RBF, it achieved the basic anticipated task of design.

• Periodical: Advanced Materials Research (Volumes 1049-1050)
• Pages: 1292-1296
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.1049-1050.1292
• Online since: October 2014
• Authors: Qing Feng Xia
• Keywords: ELM-RBF, Genetic Algorithm (GA), Optimization Strategy

References

[1] BROOMHEAD D S, LOWE D. Multivariable functional interpolation and adaptive networks[J]. Complex Systems, 1988, 2: 321-355.

[2] MOODY J E, DARKEN C J. Fast learning in networks of locally tuned processing units[J]. NeuralComputation, 1989, 1(2): 281-294.

DOI: 10.1162/neco.1989.1.2.281

[3] HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machine: theory and applications[J]. Neurocomputing, 2006, 70: 489-501.

DOI: 10.1016/j.neucom.2005.12.126

[4] HUANG G B, SIEW C K. Extreme learning machine with randomly assigned RBF kernels[J]. International Journal of Information Technology, 2005, 11(1): 16-24.

[5] Bin Li ，Liyi Bin, Xuewen Rong. Intelligent optimization strategy for ELM-RBF neural network. Journal of Shandong University(Natural Science). 2010. Vol 45, No 5. 48-52.

[6] Huang G B, Zhu Q Y, Siew C K. Extreme learning machine: theory and applications[J]. Neurocomputing, 2006, 70(1): 489-501.

DOI: 10.1016/j.neucom.2005.12.126

[7] Huang G B, Zhou H, Ding X, et al. Extreme learning machine for regression and multiclass classification[J]. Systems, Man, and Cybernetics, Part B: Cybernetics, IEEE Transactions on, 2012, 42(2): 513-529.

DOI: 10.1109/tsmcb.2011.2168604

[8] Huang G B, Siew C K. Extreme learning machine with randomly assigned RBF kernels[J]. International Journal of Information Technology, 2005, 11(1): 16-24.

[9] Huang G B, Siew C K. Extreme learning machine: RBF network case[C]/Control, Automation, Robotics and Vision Conference, 2004. ICARCV 2004 8th. IEEE, 2004, 2: 1029-1036.

DOI: 10.1109/icarcv.2004.1468985

[10] Hu K, Song A, Xia M, et al. An adaptive filtering algorithm based on genetic algorithm-backpropagation network[J]. Mathematical Problems in Engineering, 2013, (2013).

DOI: 10.1155/2013/573941

[11] Hu K, Song A G, Zhang Y C, et al. Delay-range-dependent Stability Criteria of Neural Networks with Time-varying Discrete and Distributed Delays[J]. Int J Adv Robot Syst, 2014, 11: 23.

DOI: 10.5772/53817

[12] Hu K, Song A G, Wang W L, et al. Fault detection and estimation for non-Gaussian stochastic systems with time varying delay[J]. Advances in Difference Equations, 2013, 2013(1): 1-12.

DOI: 10.1186/1687-1847-2013-22

[13] Wen X, Song A. An immune evolutionary algorithm for sphericity error evaluation[J]. International Journal of Machine Tools and Manufacture, 2004, 44(10): 1077-1084.

DOI: 10.1016/j.ijmachtools.2004.02.018

[14] Min Xia, Yingchao Zhang, Liguo Weng, Xiaoling Ye, Fashion retailing forecasting based on extreme learning machine with adaptive metrics of inputs, Knowledge-Based Systems 36 (2012) 253–259.

DOI: 10.1016/j.knosys.2012.07.002

[15] Song A, Han Y, Hu H, et al. Active perception-based haptic texture sensor[J]. Sensors and Materials, 2013, 25(1): 1-15.

[16] Zhengchao Xiong. Some problems in approximation theory with radial basis function. [D]. Fudan University. 2007(In Chinese).

[17] http: /archive. ics. uci. edu/ml.