The trade-off between regularity and stability in Tikhonov regularization

Authors:
M. Thamban Nair, Markus Hegland and Robert S. Anderssen

Journal:
Math. Comp. **66** (1997), 193-206

MSC (1991):
Primary 65R30; Secondary 65J20, 45B05

DOI:
https://doi.org/10.1090/S0025-5718-97-00811-9

MathSciNet review:
1388891

Full-text PDF

Abstract | References | Similar Articles | Additional Information

Abstract: When deriving rates of convergence for the approximations generated by the application of Tikhonov regularization to ill-posed operator equations, assumptions must be made about the nature of the stabilization (i.e., the choice of the seminorm in the Tikhonov regularization) and the regularity of the least squares solutions which one looks for. In fact, it is clear from works of Hegland, Engl and Neubauer and Natterer that, in terms of the rate of convergence, there is a trade-off between stabilization and regularity. It is this matter which is examined in this paper by means of the best-possible worst-error estimates. The results of this paper provide better estimates than those of Engl and Neubauer, and also include and extend the best possible rate derived by Natterer. The paper concludes with an application of these results to first-kind integral equations with smooth kernels.

**1.**F. de Hoog, Review of Fredholm equations of the first kind, In*Application and Numerical Solution of Integral Equations*, (R.S.Anderssen, F.de Hoog and M.A.Lucas, eds.), Sijthoff & Noordhoff, Alphen aan den Rijn, Germantown, 1980, pp. 119-134. MR**82a:45002****2.**H. W. Engl and C.W. Groetsch, eds.,*Inverse and Ill-Posed Problems*, Academic Press, London, 1987. MR**90f:00004****3.**H.W. Engl and A. Neubauer, Optimal discrepancy principles for the Tikhonov regularization of integral equations of the first kind, In*Constructive Methods for the Practical Treatment of Integral Equations*, (G.Hämmerlin and K.-H.Hoffmann, eds.), Birkhäuser, Basel, Boston, Stuttgart, 1985, pp. 120-141. CMP**19:10****4.**S. George and M.T. Nair, Parameter choice by discrepancy principles for ill-posed problems leading to optimal convergence rates,*J. Optim. Theory Appl.*,**83**(1994), 217-222. MR**95i:65094****5.**S. George and M. T. Nair, On a generalized Arcangeli's method for Tikhonov regularization with inexact data,*Research Report, CMA-MR 43-93; SMS-88-93, Australian National University*, 1993.**6.**C. W. Groetsch,*The Theory of Tikhonov Regularization for Fredholm Equations of the first Kind*, Pitman, Boston, 1984. MR**85k:45020****7.**C. W. Groetsch,*Inverse Problems in the Mathematical Sciences*, Vieweg Publishing, Wiesbaden, 1993. MR**94m:00008****8.**C. W. Groetsch and J. T. King The saturation phenomena for Tikhonov regularization.*J. Austral. Math. Soc. ( Series A )***35**(1983), 254-262. MR**84k:47011****9.**M. Hegland,*Numerische Lösung von Fredholmschen Integralgleichungen erster Art bei ungenauen Daten*. PhD thesis, ETHZ, 1988.**10.**M. Hegland, An optimal order regularization method which does not use additional smoothness assumptions,*SIAM J. Numer. Anal.***29**(1992), 1446-1461. MR**93j:65090****11.**M. Hegland, Variable Hilbert Scales and their Interpolation Inequalities with Applications to Tikhonov Regularization,*Applicable Anal.***59**(1995), 207-223. CMP**96:09****12.**T. Kato,*Perturbation Theory for Linear Operators*, 2nd Edition, Springer-Verlag, Berlin, Heidelberg, New York, 1976. MR**53:11389****13.**S. G. Krein and J. I. Petunin, Scales of Banach spaces,*Russian Math. Surveys***21**(1966), 85-160. MR**33:1719****14.**J. Locker and P. M. Prenter, Regularization with differential operators. I: General theory,*J. Math. Anal. and Appl.***74**(1980), 504-529. MR**83j:65062a****15.**J. Locker and P. M. Prenter, Regularization with differential operators. II: Weak least squares finite element solutions to first kind integral equations,*SIAM J. Numer. Anal.***17**(1980), 247-267. MR**83j:65062b****16.**J. T. Marti, Numerical solution of Fujita's equation, In*Improperly Posed Problems and Their Numerical Treatment*, Intern. Ser. Numer. Math.**63**(G.Hämmerlin and K.-H. Hoffmann, eds.), Birkhäuser, Basel, 1983, pp. 179-187. MR**86a:65134****17.**C. A. Micchelli and T. J. Rivlin, A survey of optimal recovery, In*Optimal Estimation in Approximation Theory*(C.A.Micchelli and T.J.Rivlin, eds.), Plenum Press, New York, London, 1977, pp. 1-53. MR**58:29718****18.**V. A. Morozov,*Methods of Solving Incorrectly Posed Problems*, Springer-Verlag, New York, Berlin Heidelberg, 1984. MR**86d:65005****19.**M. T. Nair, A generalization of Arcangeli's method for ill-posed problems leading to optimal rates,*Integral Equations Operator Theory***15**(1992), 1042-1046. MR**93j:65091****20.**F. Natterer, Error bounds for Tikhonov regularization in Hilbert scales,*Applicable Anal.***18**(1984), 29-37. MR**86e:65081****21.**A. Neubauer, An a posteriori parameter choice for Tikhonov regularization in Hilbert scales leading to optimal convergence rates.*SIAM J. Numer. Anal.***25**(1988), 1313-1326. MR**90b:65108****22.**D. L. Phillips, A technique for the numerical solution of certain integral equations of the first kind.*J. Assoc. Comput. Mach.***9**(1962), 84-97. MR**24:B534****23.**E. Schock, Parameter choice by discrepancy principles for the approximate solution of ill-posed problems,*Integral Equations Operator Theory***7**(1984), 895-898. MR**86m:65165****24.**E. Schock, On the asymptotic order of accuracy of Tikhonov regularizations,*J. Optim. Theory Appl.***44**(1984), 95-104. MR**86c:47012****25.**A. N. Tikhonov and V. Y. Arsenin,*Solutions of Ill-Posed Problems*, Wiley, New York, 1977. MR**56:13604****26.**M. R. Trummer, A method for solving ill-posed linear operator equations,*SIAM J. Numer. Anal.***21**(1984), 729-737. MR**85m:65051****27.**J. M. Varah, Pitfalls in the numerical solution of linear ill-posed problems,*SIAM J. Sci. Stat. Comput.***4**(1983), 164-176. MR**84g:65052**

Retrieve articles in *Mathematics of Computation of the American Mathematical Society*
with MSC (1991):
65R30,
65J20,
45B05

Retrieve articles in all journals with MSC (1991): 65R30, 65J20, 45B05

Additional Information

**M. Thamban Nair**

Affiliation:
Centre for Mathematics and Its Applications, Australian National University, Canberra ACT 0200, Australia

Address at time of publication:
Department of Mathematics, Indian Institute of Technology, Madras - 600 036, India

Email:
mtnair@acer.iitm.ernet.in

**Markus Hegland**

Affiliation:
Centre for Mathematics and Its Applications, Australian National University, Canberra ACT 0200, Australia

Address at time of publication:
Computer Sciences Laboratory, RSISE, Australian National University, Canberra ACT 0200, Australia

Email:
Markus.Hegland@anu.edu.au

**Robert S. Anderssen**

Affiliation:
Centre for Mathematics and Its Applications, Australian National University, Canberra ACT 0200, Australia

Address at time of publication:
CSIRO Division of Mathematics and Statistics, GPO Box 1965, Canberra ACT 2601, Australia

Email:
boba@cbr.dms.csiro.au

DOI:
https://doi.org/10.1090/S0025-5718-97-00811-9

Received by editor(s):
April 13, 1994

Received by editor(s) in revised form:
November 13, 1995

Article copyright:
© Copyright 1997
American Mathematical Society