A Study on Approximation Performances of General Regression Neural Network

Abstract:

In order to study the approximation performance of general regression neural networks, the structure and algorithm of general regression neural networks are first introduced. Then general regression neural networks and back propagation neural networks improved by Levenberg-Marquardt algorithm are established through programming using MATLAB language. A certain nonlinear function is taken as an example to be approximated by the two kinds of neural networks. The simulation results indicate that compared with back propagation neural networks, general regression neural networks has better approximation precision and faster convergence speed, which means it has much better approximation ability than back propagation neural networks. Therefore, for more complex function approximation, general regression neural networks is recommended. It can reduce the complexity of neural networks and it is also easier to design.

• Periodical: Applied Mechanics and Materials (Volume 441)
• Pages: 713-716
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.441.713
• Online since: December 2013
• Authors: Shuo Ding, Xiao Heng Chang, Qing Hui Wu
• Keywords: Approximation Performance, Back Propagation (BP) Neural Network, General Regression Neural Network, MATLAB, Simulation

References

[1] Rong Xie, Xinmin Wang, Yan Li and Kairui Zhao. Research and Application on Improved BP Neural Network Algorithm, 2010 5th IEEE Conference on Industrial Electronics and Application, Vol. 3 (2010), pp.1462-1466.

DOI: 10.1109/iciea.2010.5514820

[2] XING Xiao-zhe, LUO Wei. State Estimation in Power System based on Generalized Regression Neural Network Algorithm[J]. Heilongjiang Electric Power, Vol. 1 (2011), pp.50-52. (in Chinese).

[3] DING Shuo, WU Qing-hui. A MATLAB-based Study on Approximation Performances of Improved Algorithms of Typical BP Neural Networks[J]. Applied Mechanics and Materials, Vols. 313-314 (2013), pp.1353-1356.

DOI: 10.4028/www.scientific.net/amm.313-314.1353

[4] DING Shuo, CHANG Xiao-heng. A MATLAB-based Study on the Realization and Approximation Performance of RBF Neural Networks [J]. Applied Mechanics and Materials, Vols. 325-326 (2013), pp.1746-1749.

DOI: 10.4028/www.scientific.net/amm.325-326.1746