# Solving Function Approximation Problems Using the $$L^2$$-Norm of the Log Ratio as a Metric

• Ivan D. Gospodinov
• Stefan M. Filipov
• Atanas V. Atanassov
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11189)

## Abstract

This article considers the following function approximation problem: Given a non-negative function and a set of equality constraints, find the closest to it non-negative function which satisfies the constraints. As a measure of distance we propose the $$L^2$$-norm of the logarithm of the ratio of the two functions. As shown, this metric guarantees that (i) the sought function is non-negative and (ii) to the extent to which the constraints allow, the magnitude of the difference between the sought and the given function is proportional to the magnitude of the given function. To solve the problem we convert it to a finite dimensional constrained optimization problem and apply the method of Lagrange multipliers. The resulting nonlinear system, together with the system for the constraints, are solved self-consistently by applying an appropriate iterative procedure.

## Keywords

Non-negativity Relative difference Constrained optimization Lagrange multipliers

## References

1. 1.
Achiezer, N.I.: Theory of Approximation (translated by C. J. Hyman). Ungar, New York (1956)Google Scholar
2. 2.
Timan, A.F.: Theory of Approximation of Functions of a Real Variable. Pergamon (Translated from Russian) (1963)
3. 3.
Korneichuk, N.P., Ligun, A.A., Doronin, V.G.: Approximation with Constraints, Kiev (1982). (in Russian)Google Scholar
4. 4.
Milovanović, G.V., Wrigge, S.: Least squares approximation with constraints. Math. Comput. 46(174), 551–565 (1986). AMS
5. 5.
Lorentz, G. G.: Approximation of Functions. AMS (2005)Google Scholar
6. 6.
Pedregal, P.: Optimization and Approximation. UNITEXT, vol. 108. Springer, Cham (2017).
7. 7.
Gauss, C.F.: Theory of the Combination of Observations Least Subject to Errors. Society for Industrial and Applied Mathematics (Translated from original 1820 manuscript by G.W. Stewart) (1995)Google Scholar
8. 8.
Lawson, C.L., Hanson, R.J.: Solving Least Squares Problems. SIAM, Philadelphia (1995)
9. 9.
Bjorck, A.: Numerical Methods for Least Squares Problems. Society for Industrial and Applied Mathematics, Philadelphia (1996)
10. 10.
Bertsekas, D.P.: Constrained Optimization and Lagrange Multiplier Methods. Massachusetts Institute of Technology, Athena Scientific, Belmont (1996)
11. 11.
Kuhn, H. W., Tucker, A. W.: Nonlinear programming. In: Proceedings of 2nd Berkeley Symposium, Berkeley, pp. 481–492, MR 0047303. University of California Press (1951)Google Scholar
12. 12.
Karush, W.: Minima of Functions of Several Variables with Inequalities as Side Constraints. M.Sc. Dissertation, Department of Mathematics, University of Chicago, Chicago, Illinois (1939)Google Scholar
13. 13.
Luenberger, D.G., Ye, Y.: Penalty and Barrier Methods. In: Linear and Nonlinear Programming. International Series in Operations Research & Management Science, vol. 228. Springer, Cham (2016).
14. 14.
Freund, R.M.: Penalty and barrier methods for constrained optimization. Lecture Notes, Massachusetts Institute of Technology (2004)Google Scholar
15. 15.
Fukushima, M.: Solving inequality constrained optimization problems by differential homotopy continuation methods. J. Math. Anal. Appl. 133(1), 109–121 (1988)
16. 16.
Li, J., Yang, Z.: A QP-free algorithm without a penalty function or a filter for nonlinear general-constrained optimization. Appl. Math. Comp. 316, 52–72 (2018)
17. 17.
Strekalovsky, A. S., Minarchenko, I. M.: A local search method for optimization problem with d.c. inequality constraints. Appl. Math. Model (2017). https://doi.org/10.1016/j.apm.2017.07.031
18. 18.
Graff, C.: Expressing relative differences (in percent) by the difference of natural logarithms. J. Math. Psych. 60, 82–85 (2014)
19. 19.
Graff, C.: Why estimating relative differences by Ln(A/B) in percentage and why naming it geometric difference. In: Conference: ICPS, Amsterdam, Netherlands (2015). https://hal.archives-ouvertes.fr/hal-01480972/document
20. 20.
Filipov, S.M., Atanassov, A., Gospodinov, I.D.: Constrained functional similarity by minimizing the $$H^1$$ seminorm and applications to engineering problems. J. Sci. Eng. Educ. 1(1), 61–67 (2016). University of Chemical Technology and Metallurgy. http://dl.uctm.edu/see/node/jsee2016-1/12-Filipov_61-67.pdf
21. 21.
Filipov, S.M., Atanasov, A.V., Gospodinov, I.D.: Constrained relative functional similarity by minimizing the H1 semi-norm of the logarithmic difference. In: International Conference Automatics 2016, Proceedings of Technical University of Sofia, vol. 66, no. 2, pp. 349–358. Technical University of Sofia (2016). http://proceedings.tu-sofia.bg/volumes/Proceedings_volume_66_book_2_2016.pdf
22. 22.
Gospodinov, I.G., Krumov, K., Filipov, S.M.: Laplacian preserving transformation of surfaces and application to boundary value problems for Laplace’s and Poisson’s equations. Math. Model. 1(1), 14–17 (2017). STUME. https://stumejournals.com/journals/mm/2017/1/14Google Scholar
23. 23.
Filipov, S.M., Gospodinov, I.D., Faragó, I.: Shooting-projection method for two-point boundary value problems. Appl. Math. Lett. 72, 10–15 (2017).
24. 24.
Filipov, S.M., Gospodinov, I.D., Angelova, J.: Solving two-point boundary value problems for integro-differential equations using the simple shooting-projection method. In: Dimov, I., Fidanova, S., Lirkov, I. (eds.) NMA 2014. LNCS, vol. 8962, pp. 169–177. Springer, Cham (2015).

© Springer Nature Switzerland AG 2019

## Authors and Affiliations

• Ivan D. Gospodinov
• 1
Email author
• Stefan M. Filipov
• 1
• Atanas V. Atanassov
• 1
1. 1.Department of Computer ScienceUniversity of Chemical Technology and MetallurgySofiaBulgaria