Abstract
We exhibit linear problems for which every linear algorithm has infinite error, and show a (mildly) nonlinear algorithm with finite error. The error of this nonlinear algorithm can be arbitrarily small if appropriate information is used. We illustrate these examples by the inversion of a finite Laplace transform, a problem arising in remote sensing.
Similar content being viewed by others
References
Bakhvalov, N. S.,On the optimality of linear methods for operator approximation in convex classes of functions. U.S.S.R. Computational Math. and Math. Phys.11 (1971), 244–249.
Dunn, H. S.,A generalization of the Laplace transform. Proc. Cambridge Philos. Soc.63 (1967), 155–160.
Osipenko, K. Yu,Best approximation of analytic functions from information about their values at a finite number of points. Math. Notes19 (1976), 17–23.
Packel, E. W.,Linear problems (with extended range) have linear optimal algorithms. Computer Science Department Research Report CUCS-106-84, Columbia University, New York, 1984. Aequationes Math.30 (1986), 18–25.
Traub, J. F. andWoźniakowski, H.,A General theory of optimal algorithms. Academic Press, New York, 1980.
Twomey, S.,Introduction to the mathematics of inversion in remote sensing and indirect measurement. Developments in Geomathematics3, Elsevier Scientific Publ., Amsterdam, 1977.
Werschulz, A. G.,What is the complexity of ill-posed problems?. In preparation.
Werschulz, A. G.,Optimal residual algorithms. In preparation.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Werschulz, A.G., Woźniakowski, H. Are linear algorithms always good for linear problems?. Aeq. Math. 31, 202–212 (1986). https://doi.org/10.1007/BF02188189
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02188189