On the lifting of deterministic convergence rates for inverse problems with stochastic noise

Daniel Gerth; Andreas Hofinger; Ronny Ramlau

doi:10.3934/ipi.2017031

Article Contents

2017, Volume 11, Issue 4: 663-687. Doi: 10.3934/ipi.2017031

This issue Previous Article Convergence and stability of iteratively reweighted least squares for low-rank matrix recovery Next Article Non-convex TV denoising corrupted by impulse noise

On the lifting of deterministic convergence rates for inverse problems with stochastic noise

1.
Technische Universität Chemnitz, Fakultät für Mathematik, D-09107 Chemnitz, Germany
2.
Radon Institute for Computational and Applied Mathematics (RICAM), Altenbergerstraße 69, A-4040 Linz, Austria
3.
Radon Institute for Computational and Applied Mathematics (RICAM), (also Industrial Mathematics Institute, Johannes Kepler University Linz), Altenbergerstraße 69, A-4040 Linz, Austria

* Corresponding author: Daniel Gerth
* Corresponding author: Daniel Gerth

Received: March 31, 2016

Revised: March 31, 2017

Published: October 2017

The first author was supported in part by the Austrian Science Fund (FWF): W1214-N15 and by the German Research Foundation (DFG) under grants HO1454/8-2 and HO1454/10-1.

Abstract Full Text(HTML) Figure(1) Related Papers Cited by

Abstract

Both for the theoretical and practical treatment of Inverse Problems, the modeling of the noise is crucial. One either models the measurement via a deterministic worst-case error assumption or assumes a certain stochastic behavior of the noise. Although some connections between both models are known, the communities develop rather independently. In this paper we seek to bridge the gap between the deterministic and the stochastic approach and show convergence and convergence rates for Inverse Problems with stochastic noise by lifting the theory established in the deterministic setting into the stochastic one. This opens the wide field of deterministic regularization methods for stochastic problems without having to do an individual stochastic analysis for each problem.

Keywords:

Mathematics Subject Classification: Primary: 60H35, 65R32; Secondary: 47J06.

Citation:

Full Text(HTML)

Figure 1. $\mathbb{E}(||\epsilon||)^2/ \alpha$ (dashed) and regularization parameter $\alpha$ (solid) versus variance $\eta$. Left: A constant value of $\tau$ in the discrepancy principle with the expectation of the noise leads to the regularization parameter decreasing too fast, thus the deterministic condition $\delta^2/\alpha\rightarrow0$ is violated (dashed line) Right: increasing $\tau$ appropriately with decreasing variance resolves this issue

Download: Full-size image PowerPoint slide

Related Papers

Cited by

References

[1]	S. W. Anzengruber and R. Ramlau, Morozov's discrepancy principle for Tikhonov-type functionals with nonlinear operators Inverse Problems, 26 (2010), 025001, 17pp. doi: 10.1088/0266-5611/26/2/025001.
[2]	S. Birkholz, G. Steinmeyer, S. Koke, D. Gerth, S. Bürger and B. Hofmann, Phase retrieval via regularization in self-diffraction-based spectral interferometry, JOSA B, 32 (2015), 983-992. doi: 10.1364/JOSAB.32.000983.
[3]	N. Bissantz, T. Hohage and A. Munk, Consistency and rates of convergence of nonlinear {T}ikhonov regularization with random noise, Inverse Problems, 20 (2004), 1773-1789. doi: 10.1088/0266-5611/20/6/005.
[4]	N. Bissantz, T. Hohage, A. Munk and F. Ruymgaart, Convergence rates of general regularization methods for statistical inverse problems and applications, SIAM J. Numer. Anal., 45 (2007), 2610-2636. doi: 10.1137/060651884.
[5]	G. Blanchard and P. Mathé, Discrepancy principle for statistical inverse problems with application to conjugate gradient iteration, Inverse Problems, 28 (2012), 115011, 23pp. doi: 10.1088/0266-5611/28/11/115011.
[6]	V. I. Bogachev, Gaussian Measures, AMS, Providence RI, 1998. doi: 10.1090/surv/062.
[7]	D. Calvetti and E. Somersalo, An introduction to Bayesian Scientific Computing Ten Lectures on Subjective Computing, Springer, New York, 2007.
[8]	R. M. Corless, G. H. Gonnet, D. E. G. Hare, D. J. Jeffrey and D. E. Knuth, On the Lambert $W$ function, Adv. Comput. Math., 5 (1996), 329-359. doi: 10.1007/BF02124750.
[9]	I. Daubechies, M. Defrise and C. De Mol, An iterative thresholding algorithm for linear inverse problems with a sparsity constraint, Comm. Pure Appl. Math., 57 (2004), 1413-1457. doi: 10.1002/cpa.20042.
[10]	R. M. Dudley, Real Analysis and Probability Wadsworth & Brooks/Cole Advanced Books & Software, Pacific Grove, CA, 1989.
[11]	D. Th. Egoroff, Sur les suites de fonctions mesurables, CR Acad. Sci. Paris, 152 (1911), 244-246.
[12]	H. W. Engl, M. Hanke and A. Neubauer, Regularization of Inverse Problems, Kluwer Academic Publishers Group, Dordrecht, 1996. doi: 10.1007/978-94-009-1740-8.
[13]	S. N. Evans and P. B. Stark, Inverse problems as statistics, Inverse Problems, 18 (2002), R55-R97. doi: 10.1088/0266-5611/18/4/201.
[14]	K. Fan, Entfernung zweier zufälligen Grössen und die Konvergenz nach Wahrscheinlichkeit, Mathematische Zeitschrift, 49 (1944), 681-683. doi: 10.1007/BF01174225.
[15]	M. Gardner, White and brown music, fractal curves and one-over-f fluctuations, Scientific American, 238 (1978), 16-32.
[16]	D. Gerth, Problem-adapted Regularization for Inverse Problems in the Deterministic and Stochastic Setting, Ph. D thesis, Johannes Kepler University Linz, 2015.
[17]	D. Gerth, B. Hofmann, S. Birkholz, S. Koke and G. Steinmeyer, Regularization of an autoconvolution problem in ultrashort laser pulse characterization, Inverse Probl. Sci. Eng, 22 (2014), 245-266. doi: 10.1080/17415977.2013.769535.
[18]	D. Gerth and R. Ramlau, A stochastic convergence analysis for Tikhonov regularization with sparsity constraints Inverse Problems, 30 (2014), 055009, 24pp. doi: 10.1088/0266-5611/30/5/055009.
[19]	R. Gorenflo and B. Hofmann, On autoconvolution and regularization, Inverse Problems, 10 (1994), 353-373. doi: 10.1088/0266-5611/10/2/011.
[20]	M. Hanke, A. Neubauer and O. Scherzer, A convergence analysis of the Landweber iteration for nonlinear ill-posed problems, Numer. Math., 72 (1995), 21-37. doi: 10.1007/s002110050158.
[21]	T. Helin and M. Burger, Maximum a posteriori probability estimates in infinite-dimensional Bayesian inverse problems Inverse Problems, 31 (2015), 085009, 22pp. doi: 10.1088/0266-5611/31/8/085009.
[22]	A. Hofinger, Ill-posed problems: Extending the Deterministic Theory to a Stochastic Setup, Ph. D thesis, Johannes Kepler University Linz, 2006.
[23]	A. Hofinger, Ill-posed problems: Extending the Deterministic Theory to a Stochastic Setup, Trauner-Verlag, Linz, 2006.
[24]	A. Hofinger, The metrics of prokhorov and ky fan for assessing uncertainty in inverse problems, FWF Sitzungsber. Abt. Ⅱ, 215 (2006), 107–125, Available online http://www.planet-austria.at/0xc1aa500d_0x00239061.pdf
[25]	A. Hofinger and H. K. Pikkarainen, Convergence rate for the Bayesian approach to linear inverse problems, Inverse Problems, 23 (2007), 2469-2484. doi: 10.1088/0266-5611/23/6/012.
[26]	B. Hofmann, Regularization for Applied Inverse and Ill-posed Problems, Teubner Verlagsgesellschaft, Leipzig, 1986. doi: 10.1007/978-3-322-93034-7.
[27]	T. Hohage and F. Werner, Convergence rates in expectation for Tikhonov-type regularization of inverse problems with Poisson data Inverse Problems, 28 (2012), 104004, 15pp. doi: 10.1088/0266-5611/28/10/104004.
[28]	J. Kaipio and E. Somersalo, Statistical and Computational Inverse Problems, Springer, New York, 2005.
[29]	H. Kekkonen, M. Lassas and S. Siltanen, Analysis of regularized inversion of data corrupted by white Gaussian noise Inverse Problems, 30 (2014), 045009, 18pp. doi: 10.1088/0266-5611/30/4/045009.
[30]	S. Kogan, Electronic Noise and Fluctuations in Solids, Cambridge University Press, Cambridge, 1996. doi: 10.1017/CBO9780511551666.
[31]	M. Lassas, E. Saksman and S. Siltanen, Discretization-invariant Bayesian inversion and Besov space priors, Inverse Probl. Imaging, 3 (2009), 87-122. doi: 10.3934/ipi.2009.3.87.
[32]	A. K. Louis, Inverse und Schlecht Gestellte Probleme, B. G. Teubner, Stuttgart, 1989. doi: 10.1007/978-3-322-84808-6.
[33]	K. Mosegaard and M. Sambridge, Monte Carlo analysis of inverse problems, Inverse Problems, 18 (2002), R29-R54. doi: 10.1088/0266-5611/18/3/201.
[34]	A. Neubauer and H. K. Pikkarainen, Convergence results for the Bayesian inversion theory, J. Inverse Ill-Posed Probl., 16 (2008), 601-613. doi: 10.1515/JIIP.2008.032.
[35]	A. M. Stuart, Inverse Problems: A Bayesian perspective, Acta Numerica, 19 (2010), 451-559. doi: 10.1017/S0962492910000061.
[36]	A. Tarantola, Inverse Problem Theory and Methods for Model Parameter Estimation, SIAM, Philadelphia, 2005. doi: 10.1137/1.9780898717921.