Annals of Statistics

Multiscale scanning in inverse problems

Katharina Proksch, Frank Werner, and Axel Munk

Full-text: Access denied (no subscription detected)

We're sorry, but we are unable to provide you with the full text of this article because we are not able to identify you as a subscriber. If you have a personal subscription to this journal, then please login. If you are already logged in, then you may need to update your profile to register your subscription. Read more about accessing full-text


In this paper, we propose a multiscale scanning method to determine active components of a quantity $f$ w.r.t. a dictionary $\mathcal{U}$ from observations $Y$ in an inverse regression model $Y=Tf+\xi$ with linear operator $T$ and general random error $\xi$. To this end, we provide uniform confidence statements for the coefficients $\langle\varphi,f\rangle$, $\varphi\in\mathcal{U}$, under the assumption that $(T^{*})^{-1}(\mathcal{U})$ is of wavelet-type. Based on this, we obtain a multiple test that allows to identify the active components of $\mathcal{U}$, that is, $\langle f,\varphi\rangle\neq0$, $\varphi\in\mathcal{U}$, at controlled, family-wise error rate. Our results rely on a Gaussian approximation of the underlying multiscale statistic with a novel scale penalty adapted to the ill-posedness of the problem. The scale penalty furthermore ensures convergence of the statistic’s distribution towards a Gumbel limit under reasonable assumptions. The important special cases of tomography and deconvolution are discussed in detail. Further, the regression case, when $T=\text{id}$ and the dictionary consists of moving windows of various sizes (scales), is included, generalizing previous results for this setting. We show that our method obeys an oracle optimality, that is, it attains the same asymptotic power as a single-scale testing procedure at the correct scale. Simulations support our theory and we illustrate the potential of the method as an inferential tool for imaging. As a particular application, we discuss super-resolution microscopy and analyze experimental STED data to locate single DNA origami.

Article information

Ann. Statist., Volume 46, Number 6B (2018), 3569-3602.

Received: June 2017
Revised: October 2017
First available in Project Euclid: 11 September 2018

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Primary: 62G10: Hypothesis testing
Secondary: 62G15: Tolerance and confidence regions 62G20: Asymptotic properties 62G32: Statistics of extreme values; tail inference

Multiscale analysis scan statistic ill-posed problem deconvolution super-resolution Gumbel extreme value limit


Proksch, Katharina; Werner, Frank; Munk, Axel. Multiscale scanning in inverse problems. Ann. Statist. 46 (2018), no. 6B, 3569--3602. doi:10.1214/17-AOS1669.

Export citation


  • Abramovich, F. and Silverman, B. W. (1998). Wavelet decomposition approaches to statistical inverse problems. Biometrika 85 115–129.
  • Albani, V., Elbau, P., de Hoop, M. V. and Scherzer, O. (2016). Optimal convergence rates results for linear inverse problems in Hilbert spaces. Numer. Funct. Anal. Optim. 37 521–540.
  • Anderssen, R. S. (1986). The linear functional strategy for improperly posed problems. In Inverse Problems (Oberwolfach, 1986). Internat. Schriftenreihe Numer. Math. 77 11–30. Birkhäuser, Basel.
  • Arias-Castro, E., Donoho, D. L. and Huo, X. (2005). Near-optimal detection of geometric objects by fast multiscale methods. IEEE Trans. Inform. Theory 51 2402–2425.
  • Aspelmeier, T., Egner, A. and Munk, A. (2015). Modern statistical challenges in high-resolution fluorescence microscopy. Annu. Rev. Stat. Appl. 2 163–202.
  • Bertero, M., Boccacci, P., Desiderà, G. and Vicidomini, G. (2009). Image deblurring with Poisson data: From cells to galaxies. Inverse Probl. 25 025004.
  • Bissantz, N., Hohage, T., Munk, A. and Ruymgaart, F. (2007). Convergence rates of general regularization methods for statistical inverse problems and applications. SIAM J. Numer. Anal. 45 2610–2636.
  • Bissantz, N., Claeskens, G., Holzmann, H. and Munk, A. (2009). Testing for lack of fit in inverse regression—With applications to biophotonic imaging. J. R. Stat. Soc. Ser. B. Stat. Methodol. 71 25–48.
  • Burger, M., Flemming, J. and Hofmann, B. (2013). Convergence rates in regularization if the sparsity assumption fails. Inverse Probl. 29 025013.
  • Butucea, C. (2007). Goodness-of-fit testing and quadratic functional estimation from indirect observations. Ann. Statist. 35 1907–1930.
  • Butucea, C. and Comte, F. (2009). Adaptive estimation of linear functionals in the convolution model and applications. Bernoulli 15 69–98.
  • Butucea, C. and Ingster, Y. I. (2013). Detection of a sparse submatrix of a high-dimensional noisy matrix. Bernoulli 19 2652–2688.
  • Castillo, I. and Nickl, R. (2014). On the Bernstein–von Mises phenomenon for nonparametric Bayes procedures. Ann. Statist. 42 1941–1969.
  • Cavalier, L. and Golubev, Y. (2006). Risk hull method and regularization by projections of ill-posed inverse problems. Ann. Statist. 34 1653–1677.
  • Cavalier, L. and Tsybakov, A. (2002). Sharp adaptation for inverse problems with random noise. Probab. Theory Related Fields 123 323–354.
  • Cavalier, L., Golubev, Y., Lepski, O. and Tsybakov, A. (2003). Block thresholding and sharp adaptive estimation in severely ill-posed inverse problems. Teor. Veroyatn. Primen. 48 534–556.
  • Chan, H. P. and Walther, G. (2013). Detection with the scan and the average likelihood ratio. Statist. Sinica 23 409–428.
  • Chernousova, E. and Golubev, Y. (2014). Spectral cut-off regularizations for ill-posed linear models. Math. Methods Statist. 23 116–131.
  • Chernozhukov, V., Chetverikov, D. and Kato, K. (2014). Gaussian approximation of suprema of empirical processes. Ann. Statist. 42 1564–1597.
  • Cohen, A., Hoffmann, M. and Reiß, M. (2004). Adaptive wavelet Galerkin methods for linear inverse problems. SIAM J. Numer. Anal. 42 1479–1501.
  • Dedecker, J., Merlevède, F. and Rio, E. (2014). Strong approximation of the empirical distribution function for absolutely regular sequences in $\mathbb{R}^{d}$. Electron. J. Probab. 19 1–56.
  • Dickhaus, T. (2014). Simultaneous Statistical Inference: With Applications in the Life Sciences. Springer, Heidelberg.
  • Donoho, D. L. (1995). Nonlinear solution of linear inverse problems by wavelet-vaguelette decomposition. Appl. Comput. Harmon. Anal. 2 101–126.
  • Dümbgen, L. and Spokoiny, V. (2001). Multiscale testing of qualitative hypotheses. Ann. Statist. 29 124–152.
  • Dümbgen, L. and Walther, G. (2008). Multiscale inference about a density. Ann. Statist. 36 1758–1785.
  • Eckle, K., Bissantz, N. and Dette, H. (2017). Multiscale inference for multivariate deconvolution. Electron. J. Stat. 11 4179–4219.
  • Eckle, K., Bissantz, N., Dette, H., Proksch, K. and Einecke, S. (2018). Multiscale inference for a multivariate density with applications to X-ray astronomy. Ann. Inst. Statist. Math. To appear. DOI:10.1007/s10463-017-0605-1.
  • Fan, J. (1991). Asymptotic normality for deconvolution kernel density estimators. Sankhyā, Ser. A 53 97–110.
  • Friedenberg, D. A. and Genovese, C. R. (2013). Straight to the source: Detecting aggregate objects in astronomical images with proper error control. J. Amer. Statist. Assoc. 108 456–468.
  • Genovese, C. R., Perone-Pacifico, M., Verdinelli, I. and Wasserman, L. (2012). The geometry of nonparametric filament estimation. J. Amer. Statist. Assoc. 107 788–799.
  • Goldenshluger, A. (1999). On pointwise adaptive nonparametric deconvolution. Bernoulli 5 907–925.
  • Hell, S. (2007). Far-field optical nanoscopy. Science 316 1153–1158.
  • Hell, S. W. and Wichmann, J. (1994). Breaking the diffraction resolution limit by stimulated emission: Stimulated-emission-depletion fluorescence microscopy. Opt. Lett. 19 780–782.
  • Hohage, T. and Werner, F. (2016). Inverse problems with Poisson data: Statistical regularization theory, applications and algorithms. Inverse Probl. 32 093001.
  • Holzmann, H., Bissantz, N. and Munk, A. (2007). Density testing in a contaminated sample. J. Multivariate Anal. 98 57–75.
  • Ingster, Y. I. (1993). Asymptotically minimax hypothesis testing for nonparametric alternatives. I–III. Math. Methods Statist. 2 85–114, 171–189, 249–268.
  • Ingster, Y., Laurent, B. and Marteau, C. (2014). Signal detection for inverse problems in a multidimensional framework. Math. Methods Statist. 23 279–305.
  • Ingster, Y. I., Sapatinas, T. and Suslina, I. A. (2012). Minimax signal detection in ill-posed inverse problems. Ann. Statist. 40 1524–1549.
  • Johnstone, I. M. and Paul, D. (2014). Adaptation in some linear inverse problems. Stat 3 187–199.
  • Johnstone, I. M. and Silverman, B. W. (1991). Discretization effects in statistical inverse problems. J. Complexity 7 1–34.
  • Johnstone, I. M., Kerkyacharian, G., Picard, D. and Raimondo, M. (2004). Wavelet deconvolution in a periodic setting. J. R. Stat. Soc. Ser. B. Stat. Methodol. 66 547–573.
  • Kabluchko, Z. (2011). Extremes of the standardized Gaussian noise. Stochastic Process. Appl. 121 515–533.
  • Kazantsev, I., Lemahieu, I., Salov, G. and Denys, R. (2002). Statistical detection of defects in radiographic images in nondestructive testing. Signal Process. 82 791–801.
  • Kerkyacharian, G., Kyriazis, G., Le Pennec, E., Petrushev, P. and Picard, D. (2010). Inversion of noisy Radon transform by SVD based needlets. Appl. Comput. Harmon. Anal. 28 24–45.
  • Klar, T. A. and Hell, S. W. (1999). Subdiffraction resolution in far-field fluorescence microscopy. Opt. Lett. 24 954–956.
  • Knapik, B. T., van der Vaart, A. W. and van Zanten, J. H. (2011). Bayesian inverse problems with Gaussian priors. Ann. Statist. 39 2626–2657.
  • Komlós, J., Major, P. and Tusnády, G. (1975). An approximation of partial sums of independent RV’s and the sample DF. I. Z. Wahrsch. Verw. Gebiete 32 111–131.
  • Kou, J. (2017). Identifying the support of rectangular signals in Gaussian noise. Preprint. Available at arXiv:1703.06226.
  • Laurent, B., Loubes, J.-M. and Marteau, C. (2011). Testing inverse problems: A direct or an indirect problem? J. Statist. Plann. Inference 141 1849–1861.
  • Laurent, B., Loubes, J.-M. and Marteau, C. (2012). Non asymptotic minimax rates of testing in signal detection with heterogeneous variances. Electron. J. Stat. 6 91–122.
  • Li, H., Munk, A., Sieling, H. and Walther, G. (2016). The essential histogram. Preprint. Available at arXiv:1612.07216.
  • Lin, G. D. (2017). Recent developments on the moment problem. Preprint. Available at arXiv:1703:01027.
  • Mair, B. A. and Ruymgaart, F. H. (1996). Statistical inverse estimation in Hilbert scales. SIAM J. Appl. Math. 56 1424–1444.
  • Marteau, C. and Mathé, P. (2014). General regularization schemes for signal detection in inverse problems. Math. Methods Statist. 23 176–200.
  • Mathé, P. and Pereverzev, S. V. (2002). Direct estimation of linear functionals from indirect noisy observations. J. Complexity 18 500–516.
  • Meister, A. (2009). Deconvolution Problems in Nonparametric Statistics. Lecture Notes in Statistics 193. Springer, Berlin.
  • Natterer, F. (1986). The Mathematics of Computerized Tomography. B. G. Teubner, Stuttgart.
  • Nickl, R. and Reiß, M. (2012). A Donsker theorem for Lévy measures. J. Funct. Anal. 263 3306–3332.
  • Nikol’skiĭ, S. M. (1951). Inequalities for entire functions of finite degree and their application in the theory of differentiable functions of several variables. In Trudy Mat. Inst. Steklov. 38 244–278. Izdat. Akad. Nauk SSSR, Moscow.
  • O’Sullivan, F. (1986). A statistical perspective on ill-posed inverse problems. Statist. Sci. 1 502–527.
  • Pickands, J. III (1969). Upcrossing probabilities for stationary Gaussian processes. Trans. Amer. Math. Soc. 145 51–73.
  • Piterbarg, V. I. (1996). Asymptotic Methods in the Theory of Gaussian Processes and Fields. Translations of Mathematical Monographs 148. Amer. Math. Soc., Providence, RI.
  • Proksch, K., Werner, F. and Munk, A. (2018). Supplement to “Multiscale scanning in inverse problems.” DOI:10.1214/17-AOS1669SUPP.
  • Ray, K. (2013). Bayesian inverse problems with non-conjugate priors. Electron. J. Stat. 7 2516–2549.
  • Ray, K. (2017). Adaptive Bernstein–von Mises theorems in Gaussian white noise. Ann. Statist. 45 2511–2536.
  • Rio, E. (1993). Strong approximation for set-indexed partial-sum processes, via KMT constructions. II. Ann. Probab. 21 1706–1727.
  • Rohde, A. (2008). Adaptive goodness-of-fit tests based on signed ranks. Ann. Statist. 36 1346–1374.
  • Rufibach, K. and Walther, G. (2010). The block criterion for multiscale inference about a density, with applications to other multiscale problems. J. Comput. Graph. Statist. 19 175–190.
  • Schmidt-Hieber, J., Munk, A. and Dümbgen, L. (2013). Multiscale methods for shape constraints in deconvolution: Confidence statements for qualitative features. Ann. Statist. 41 1299–1328.
  • Schwartzman, A., Dougherty, R. F. and Taylor, J. E. (2008). False discovery rate analysis of brain diffusion direction maps. Ann. Appl. Stat. 2 153–175.
  • Sharpnack, J. and Arias-Castro, E. (2016). Exact asymptotics for the scan statistic and fast alternatives. Electron. J. Stat. 10 2641–2684.
  • Söhl, J. and Trabs, M. (2012). A uniform central limit theorem and efficiency for deconvolution estimators. Electron. J. Stat. 6 2486–2518.
  • Ta, H., Keller, J., Haltmeier, M., Saka, S. K., Schmied, J., Opazo, F., Tinnefeld, P., Munk, A. and Hell, S. W. (2015). Mapping molecules in scanning far-field fluorescence nanoscopy. Nat. Commun. 6 7977.
  • Tsybakov, A. (2000). On the best rate of adaptive estimation in some inverse problems. C. R. Acad. Sci. Paris Sér. I Math. 330 835–840.
  • Tsybakov, A. B. (2009). Introduction to Nonparametric Estimation. Springer, New York.
  • Walther, G. (2010). Optimal and fast detection of spatial clusters with scan statistics. Ann. Statist. 38 1010–1033.
  • Willer, T. (2009). Optimal bounds for inverse problems with Jacobi-type eigenfunctions. Statist. Sinica 19 785–800.

Supplemental materials

  • Supplement to “Multiscale scanning in inverse problems”. This supplementary material contains an explanation of the full width at half maximum (FWHM), a detailed mathematical model for super-resolution STED microscopy, a detailed simulation study for $2$-dimensional support inference and detailed proofs of all theoretical results provided in the main document.