Abstract
We compare the average errors of n-term linear and nonlinear approximations assuming that the coefficients in an orthogonal expansion of the approximated element are scaled i.i.d. random variables. We show that generally the n-term nonlinear approximation can be even exponentially better than the n-term linear approximation. On the other hand, if the scaling parameters decay no faster than polynomially then the average errors of nonlinear approximations do not converge to zero faster than those of linear approximations, as nāāā+ā. The main motivation and application is the approximation of Gaussian processes. In this particular case, the nonlinear approximation is, roughly, no more than n times better than its linear counterpart.
Dedicated to Ian H. Sloan on the occasion of his 80th birthday.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The word āadaptiveā here means that the removed components depend on x.
- 2.
For two positive sequences, we write a nāāāb n iff there are 0ā<ācā<āCā<ā+ā and n 0 such that cāā¤āa nāb nāā¤āC holds for all nāā„ān 0. We write a nāāāb n iff limnā+āa nāb nā=ā1.
References
Cioica, P.A., Dahlke, S., Dƶhring, N., Kinzel, S., Lindner, F., Raasch, T., Ritter, K., Schilling, R.L.: Adaptive wavelet methods for the stochastic Poisson equation. BIT Numer. Math. 52, 589ā614 (2012)
Cohen, A., DāAles, J.-P.: Nonlinear approximation of random functions. SIAM J. Appl. Math. 57, 518ā540 (1997)
Cohen, A., Daubechies, I., Guleryuz, O.G., Orchard, M.T.: On the importance of combining wavelet-based nonlinear approximation with coding strategies. IEEE Trans. Inf. Theory 48, 1895ā1921 (2002)
Creutzig, J., MĆ¼ller-Gronbach, T., Ritter, K.: Free-knot spline approximation of stochastic processes. J. Complexity 23, 867ā889 (2007)
DeVore, R.A.: Nonlinear approximation. Acta Numer. 8, 51ā150 (1998)
DeVore, R.A., Jawerth, B.: Optimal nonlinear approximation. Manuscripta Math. 63, 469ā478 (1992)
Donoho, D.L.: Compressed sensing. IEEE Trans. Inf. Theory 52, 1289ā1306 (2006)
Kon, M.A., Plaskota, L.: Information complexity of neural networks. Neural Netw. 13, 365ā376 (2000)
Kon, M.A., Plaskota, L.: Information-based nonlinear approximation: an average case setting. J. Complexity 21, 211ā229 (2005)
Kuo, H.-H.: Gaussian Measures in Banach Spaces. Lecture Notes in Mathematics, vol. 463. Springer, New York (1975)
Plaskota, L.: Noisy Information and Computational Complexity. Cambridge University Press, Cambridge (1996)
Rauhut, H.: Compressive sensing and structured random matrices. In: Fornasier, M. (ed.) Theoretical Foundations and Numerical Methods for Sparse Recovery, vol. 9, pp. 1ā92. Walter de Gruyter, Berlin (2012)
Ritter, K.: Average-Case Analysis of Numerical Problems. Springer, Berlin (2000)
Temlyakov, V.N.: Nonlinear methods of approximation. Found. Comput. Math. 3, 33ā107 (2003)
Traub, J.F., Wasilkowski, G.W., WoÅŗniakowski, H.: Information-Based Complexity. Academic Press, New York (1988)
Vakhania, N.N., Tarieladze, V.I., Chobanyan, S.A.: Probability Distributions on Banach Spaces. Kluwer, Dordrecht (1987)
Vybiral, J.: Average best m-term approximation. Constr. Approx. 36, 83ā115 (2012)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
Ā© 2018 Springer International Publishing AG, part of Springer Nature
About this chapter
Cite this chapter
Plaskota, L. (2018). On Linear Versus Nonlinear Approximation in the Average Case Setting. In: Dick, J., Kuo, F., WoÅŗniakowski, H. (eds) Contemporary Computational Mathematics - A Celebration of the 80th Birthday of Ian Sloan. Springer, Cham. https://doi.org/10.1007/978-3-319-72456-0_46
Download citation
DOI: https://doi.org/10.1007/978-3-319-72456-0_46
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-72455-3
Online ISBN: 978-3-319-72456-0
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)