DOI QR코드

DOI QR Code

A SIMULTANEOUS NEURAL NETWORK APPROXIMATION WITH THE SQUASHING FUNCTION

  • Hahm, Nahm-Woo (Department of Mathematics University of Incheon) ;
  • Hong, Bum-Il (Department of Mathematics Kyung Hee University)
  • Received : 2009.05.19
  • Accepted : 2009.06.01
  • Published : 2009.06.25

Abstract

In this paper, we actually construct the simultaneous approximation by neural networks to a differentiable function. To do this, we first construct a polynomial approximation using the Fejer sum and then a simultaneous neural network approximation with the squashing activation function. We also give numerical results to support our theory.

References

  1. B. Gao and Y. Xu, Univariant approximation by superpositions of a sigmoidal function, J. Math. Anal. Appl. 178(1993), 221-226 https://doi.org/10.1006/jmaa.1993.1302
  2. N. Hahm and B. I. Hong. Approximation order to a function in $C^1[0,1]$ and its derivative by a feedforward neural network, J. of Appl. Math. and Info. 27(2009), 137-147.
  3. N. Hahm and B. I. Hong, Approximation order to a function in $L_{p}$ space by generalized translation netowrks, Honam Math. Jour. 28(2006), 125-133.
  4. B. I. Hong and N. Hahm, Approximation order to a function in ${\bar{C}}$(R) by supersposition of a sigmoidal function, Appl. Math. Lett. 15(2002), 591-597. https://doi.org/10.1016/S0893-9659(02)80011-8
  5. B. L. Kalman and S. C. Kwasny, Why Tanh : Choosing a sigmoidal function, Int. Joint Conf. Oil Neural Networks 4(1992), 578-581.
  6. 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
  7. X. Li, Simultaneous approximation of a multivariate junctions and their derivatives by neural networks with one hidden layer, Neurocomputing 12(1996), 327-343 https://doi.org/10.1016/0925-2312(95)00070-4
  8. H. N. Mhaskar and N. Hahm, Neural networks for functional approximation and system identification, Neural Comp. 9(1997), 143-159. https://doi.org/10.1162/neco.1997.9.1.143
  9. I. P. Natanson, Constructive function theory, Frederick Publ., 1964.
  10. E. M. Stein, Singular integrals and differentiability properties of functions, Princeton Univ. Press, 1970.

Cited by

  1. THE CAPABILITY OF LOCALIZED NEURAL NETWORK APPROXIMATION vol.35, pp.4, 2013, https://doi.org/10.5831/HMJ.2013.35.4.729