CONSTRUCTIVE APPROXIMATION BY GAUSSIAN NEURAL NETWORKS

Hahm, Nahm-Woo;Hong, Bum-Il

• Accepted : 2012.06.05
• Published : 2012.09.25
• 26 7

Abstract

In this paper, we discuss a constructive approximation by Gaussian neural networks. We show that it is possible to construct Gaussian neural networks with integer weights that approximate arbitrarily well for functions in $C_c(\mathbb{R}^s)$. We demonstrate numerical experiments to support our theoretical results.

Keywords

Constructive Approximation;Neural Network;Gaussian Activation Function

References

1. L. Beheral, M. Gopal and S. Chaudhury, Trajectory tracking of robot manipulator using Gaussian networks, Robot. Auton. Syst. 13(2)(1994), 107-115. https://doi.org/10.1016/0921-8890(94)90053-1
2. D. Chen, Degree of approximation by superpositions of a sigmoidal function, Approx. Theory and Appl. 9(3)(1993), 17-28.
3. G. Cybenko, Approximation by superpositions of sigmoidal functions, Math. Control Signal 2(1989), 303-314. https://doi.org/10.1007/BF02551274
4. C. Firmin, D Hamad, J. Postaire and R. Zhang, Gaussian neural networks for glass bottles inspection : a learning procedure, Int. J. Neural Syst. 8(1)(1997), 41-46. https://doi.org/10.1142/S0129065797000069
5. E. J. Hartman, J. D. Keeler and J. M. Kowalski, Layered neural networks with Gaussian hidden units as universal approximations, Neural Comput. 2(2)(1990), 210-215. https://doi.org/10.1162/neco.1990.2.2.210
6. B. I. Hong and N. Hahm, Approximation order to a function in ${\overline{C}({\mathbb{R}})$ by superposition of a sigmoidal function, Appl. Math. Lett. 15(2002), 591-597. https://doi.org/10.1016/S0893-9659(02)80011-8
7. G. Lewicki and G. Marino, Approximation of functions of finite variation by superpositions of a sigmoidal function, Appl. Math. Lett. 17(2004), 1147-1152. https://doi.org/10.1016/j.aml.2003.11.006
8. W. Light, Techniques for generating approximations via convolution kernels, Numer. Algorithms 5(1993), 247-261. https://doi.org/10.1007/BF02210385
9. H. N. Mhaskar, Versatile Gaussian networks, Proc. IEEEWorkshop on Nonlinear Image and Signal Proc. (1995), 70-73.
10. M. A. Sartori and P. J. Antsaklis, Gaussian neural networks for control function implementation, Math. Comput. Model. 23(1996), 129-142. https://doi.org/10.1016/0895-7177(95)00223-5

Cited by

1. CONSTRUCTIVE APPROXIMATION BY NEURAL NETWORKS WITH POSITIVE INTEGER WEIGHTS vol.23, pp.3, 2015, https://doi.org/10.11568/kjm.2015.23.3.327

Acknowledgement

Supported by : University of Incheon