Skip to main content
Log in

A statistical view of iterative methods for linear inverse problems

  • Original Paper
  • Published:
TEST Aims and scope Submit manuscript

Abstract

In this article, we study the problem of recovering the unknown solution of a linear ill-posed problem, via iterative regularization methods, from a statistical point of view. The basic purpose of the paper is to develop adaptive model selection techniques for determining the regularization parameters, i.e., the iteration index. We assume observations are taken over a fixed grid and we consider solutions over a sequence of finite-dimensional subspaces. Based on concentration inequalities techniques, we derive non-asymptotic optimal upper bounds for the mean square error of the proposed estimator.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  • Baraud Y (2000) Model selection for regression on a fixed design. Probab Theory Relat Fields 117:467–493

    Article  MATH  MathSciNet  Google Scholar 

  • Birgé L, Massart P (1998) Minimum contrast estimators on sieves: exponential bounds and rates of convergence. Bernoulli 4:329–395

    Article  MATH  MathSciNet  Google Scholar 

  • Bousquet O (2003) Concentration inequalities for sub-additive functions using of entropy method. Stoch Inequal Appl 56:213–247

    MathSciNet  Google Scholar 

  • Calveltti D, Hansen P, Reichel L (2002) L-curve curvature bounds via Lanczos bidiagonalization. Electron Trans Numer Anal 14:20–35

    MathSciNet  Google Scholar 

  • Cavalier L, Golubev G, Picard D, Tsybakov A (2002) Oracle inequalities for inverse problem. Ann Stat 30(3):843–874

    Article  MATH  MathSciNet  Google Scholar 

  • Dey AK (1996) Cross-validation for parameter selection in inverse estimation problems. Scand J Stat 23(4):609–620

    MATH  MathSciNet  Google Scholar 

  • Egger H (2005) Accelerated Newton–Landweber iterations for regularizing nonlinear inverse problems. Technical Report SBF F013, RICAM, Austria

  • Engl H, Grever W (1994) Using the L-curve for determining optimal regularization parameters. Num Math 69:25–31

    Article  MATH  MathSciNet  Google Scholar 

  • Engl HW, Hanke M, Neubauer A (1996) Reguralization of inverse problems. Mathematics and its applications, vol 375. Kluwer Academic, Dordrecht

    Google Scholar 

  • Gilyazov S, Gol’dman N (2000) Reguralization of ill-posed problems by iteration methods. Mathematics and its applications, vol 499. Kluwer Academic, Dordrecht

    Google Scholar 

  • Hansen PC, O’Leary DP (1993) The use of the L-curve in the regularization of discrete ill-posed problems. SIAM J Sci Comput 14:1487–1503

    Article  MATH  MathSciNet  Google Scholar 

  • Kilmer ME, O’Leary DP (2001) Choosing regularization parameters in iterative methods for ill-posed problems. SIAM J Matrix Anal Appl 22(4):1204–1221

    Article  MATH  MathSciNet  Google Scholar 

  • Loubes JM, Ludeña C (2004) Estimators for nonlinear inverse problems. Preprints. Available at http://arxiv.org/pdf/math.ST/0502033

  • Morozov VA (1966) On the solution of functional equations by the method of regularization. Sov Math Dokl 7:414–417

    MATH  Google Scholar 

  • O’Sullivan F (1986) A statistical perspective on ill-posed inverse problems. Stat Sci 1(4):502–527

    Article  MathSciNet  Google Scholar 

  • Tikhonov A, Arsenin V (1977) Solutions of ill-posed problems. Wiley, New York

    MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Ana K. Fermín.

Additional information

The authors would like to thank Projects Agenda Petróleo, Fonacit and Ecos-Nord for their financial support.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Fermín, A.K., Ludeña, C. A statistical view of iterative methods for linear inverse problems. TEST 17, 381–400 (2008). https://doi.org/10.1007/s11749-006-0038-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11749-006-0038-2

Keywords

Mathematics Subject Classification (2000)

Navigation