Abstract
A special class of approximations of continuous functions of several variables on the unit coordinate cube is investigated. The class is constructed using Kolmogorov’s theorem stating that functions of the indicated type can be represented as a finite superposition of continuous single-variable functions and another result on the approximation of such functions by linear combinations of quadratic exponentials (also known as Gaussian functions). The effectiveness of such a representation is based on the author’s previously proved assertion that the Mexican hat mother wavelet on any fixed bounded interval can be approximated as accurately as desired by a linear combination of two Gaussian functions. It is proved that the class of approximations under study is dense everywhere in the class of continuous multivariable functions on the coordinate cube. For the case of continuous functions of two variables, numerical results are presented that confirm the effectiveness of approximations of the studied class.
Similar content being viewed by others
REFERENCES
K. L. Teo, C. J. Goh, and K. H. Wong, A Unified Computational Approach to Optimal Control Problems (Longman Scientific and Technical, New York, 1991).
A. V. Chernov, “On approximate solution of free-time optimal control problems,” Vestn. Nizhegorod. Gos. Univ. im. N.I. Lobachevskogo, No. 6 (1), 107–114 (2012).
A. V. Chernov, “Smooth finite-dimensional approximations of distributed optimization problems via control discretization,” Comput. Math. Math. Phys. 53 (12), 1839–1852 (2013).
A. V. Chernov, “On applicability of control parametrization technique to solving distributed optimization problems,” Vestn. Udmurt. Univ. Mat. Mekh. Komp’yut. Nauki 24 (1), 102–117 (2014).
A. V. Chernov, “On the smoothness of an approximated optimization problem for a Goursat–Darboux system on a varied domain,” Tr. Inst. Mat. Mekh. Ural. Otd. Ross. Akad. Nauk 20 (1), 305–321 (2014).
A. V. Chernov, “On piecewise constant approximation in distributed optimization problems,” Tr. Inst. Mat. Mekh. Ural. Otd. Ross. Akad. Nauk 21 (1), 264–279 (2015).
A. N. Kolmogorov, “On the representation of continuous functions of several variables by superposition of continuous functions of one variable and addition,” Dokl. Akad. Nauk SSSR 114 (5), 953–956 (1957).
D. A. Sprecher, “On the structure of continuous functions of several variables,” Trans. Am. Math. Soc. 115, 340–355 (1965).
A. Yu. Golubkov, “The tracing of external and internal representation functions of continuous functions of several variables by superposition of continuous functions of one variable,” Fundam. Prikl. Mat. 8 (1), 27–38 (2002).
E. Yu. Butyrskii, I. A. Kuvaldin, and V. P. Chalkin, “Approximation of multidimensional functions,” Nauchn. Priborostr. 20 (2), 82–92 (2010).
R. Hecht-Nielsen, “Kolmogorov’s mapping neural network existence theorem,” in Proceedings of the First IEEE Annual International Joint Conference on Neural Networks, San-Diego (New York, NY, USA, 1987), Vol. 3, pp. 11–13.
D. V. Alexeev, “Neural network approximation of several variable functions,” J. Math. Sci. 168 (1), 5–13 (2010).
A. N. Gorban’, “A generalized approximation theorem and an exact representation of polynomials of several variables by superpositions of polynomials of one variable,” Russ. Math. 42 (5), 4–7 (1998).
V. E. Ismailov, “Approximation by sums of ridge functions with fixed directions,” St. Petersburg Math. J. 28 (6), 741–772 (2017).
Yu. N. Kul’chin, I. V. Denisov, A. V. Panov, and N. A. Rybal’chenko, “Application of perceptrons for nonlinear reconstruction tomography,” Probl. Upr., No. 4, 59–63 (2006).
A. V. Chernov, “On using Gaussian functions with varied parameters for approximation of functions of one variable on a finite segment,” Vestn. Udmurt. Univ. Mat. Mekh. Komp’yut. Nauki 27 (2), 267–282 (2017).
A. V. Chernov, “On the application of Gaussian functions for discretization of optimal control problems,” Vestn. Udmurt. Univ. Mat. Mekh. Komp’yut. Nauki 27 (4), 558–575 (2017).
V. Maz’ya and G. Schmidt, Approximate Approximations (Am. Math. Soc., Providence, R.I., 2007).
S. D. Riemenschneider and N. Sivakumar, “Cardinal interpolation by Gaussian functions: A survey,” J. Anal. 8, 157–178 (2000).
Luh Lin-Tian, “The shape parameter in the Gaussian function,” Comput. Math. Appl. 63, 687–694 (2012).
T. Hangelbroek, W. Madych, F. Narcowich, and J. D. Ward, “Cardinal interpolation with Gaussian kernels,” J. Fourier Anal. Appl. 18 (1), 67–86 (2012).
K. Hamm, “Approximation rates for interpolation of Sobolev functions via Gaussians and allied functions,” J. Approx. Theory 189, 101–122 (2015).
M. Griebel, M. Schneider, and C. Zenger, “A combination technique for the solution of sparse grid problems,” Proceedings of the IMACS International Symposium on Iterative Methods in Linear Algebra, Belgium, April 2–4,1991 (North-Holland, Amsterdam, 1992), pp. 263–281.
E. Georgoulis, J. Levesley, and F. Subhan, “Multilevel sparse kernel-based interpolation,” SIAM J. Sci. Comput. 35 (2), A815–A831 (2013).
J. J. More and D. J. Thuente, “Line search algorithms with guaranteed sufficient decrease,” ACM Trans. Math. Software 20, 286–307 (1994).
Author information
Authors and Affiliations
Corresponding author
Additional information
Translated by I. Ruzanova
Rights and permissions
About this article
Cite this article
Chernov, A.V. Gaussian Functions Combined with Kolmogorov’s Theorem as Applied to Approximation of Functions of Several Variables. Comput. Math. and Math. Phys. 60, 766–782 (2020). https://doi.org/10.1134/S0965542520050073
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0965542520050073