October 2017, 11(5): 857-874. doi: 10.3934/ipi.2017040

Well-posed Bayesian inverse problems and heavy-tailed stable quasi-Banach space priors

Free University of Berlin and Zuse Institute Berlin, Takustraße 7,14195 Berlin, Germany

Received  May 2016 Revised  November 2016 Published  July 2017

This article extends the framework of Bayesian inverse problems in infinite-dimensional parameter spaces, as advocated by Stuart (Acta Numer. 19:451–559,2010) and others, to the case of a heavy-tailed prior measure in the family of stable distributions, such as an infinite-dimensional Cauchy distribution, for which polynomial moments are infinite or undefined. It is shown that analogues of the Karhunen–Loéve expansion for square-integrable random variables can be used to sample such measures on quasi-Banach spaces. Furthermore, under weaker regularity assumptions than those used to date, the Bayesian posterior measure is shown to depend Lipschitz continuously in the Hellinger metric upon perturbations of the misfit function and observed data.

Citation: T. J. Sullivan. Well-posed Bayesian inverse problems and heavy-tailed stable quasi-Banach space priors. Inverse Problems & Imaging, 2017, 11 (5) : 857-874. doi: 10.3934/ipi.2017040
References:
[1]

A. AchimP. Tsakalides and A. Bezerianos, SAR image denoising via Bayesian wavelet shrinkage based on heavy-tailed modeling, IEEE Trans. Geosci. Remote, 41 (2003), 1773-1784. doi: 10.1109/TGRS.2003.813488.

[2]

C. D. Aliprantis and K. C. Border, Infinite Dimensional Analysis: A Hitchhiker's Guide Springer, Berlin, third edition, 2006. doi: 10.1007/3-540-29587-9.

[3]

V. I. Bogachev, Differentiable Measures and the Malliavin Calculus volume 164 of Mathematical Surveys and Monographs, American Mathematical Society, Providence, RI, 2010. doi: 10.1090/surv/164.

[4]

R. Bonic, Some properties of Hilbert scales, Proc. Amer. Math. Soc., 18 (1967), 1000-1003. doi: 10.1090/S0002-9939-1967-0230115-3.

[5]

J. M. ChambersC. L. Mallows and B. W. Stuck, A method for simulating stable random variables, J. Amer. Statist. Assoc., 71 (1976), 340-344. doi: 10.1080/01621459.1976.10480344.

[6]

O. Christensen and D. T. Stoeva, p-frames in separable Banach spaces, Adv. Comput. Math., 18 (2003), 117-126. doi: 10.1023/A:1021364413257.

[7]

M. Dashti and A. M. Stuart, The Bayesian approach to inverse problems, Handbook of Uncertainty Quantification, (2016), 311-428. doi: 10.1007/978-3-319-11259-6_7-1.

[8]

M. DashtiS. Harris and A. M. Stuart, Besov priors for Bayesian inverse problems, Inverse Probl. Imaging, 6 (2012), 183-200. doi: 10.3934/ipi.2012.6.183.

[9]

N. Hansen, F. Gemperle, A. Auger and P. Koumoutsakos, When do heavy-tail distributions help?, In T. P. Runarsson, H. -G. Beyer, E. Burke, J. J. Merelo-Guervós, L. D. Whitley, and X. Yao, editors, Parallel Problem Solving from Nature — PPSN Ⅸ: 9th International Conference, Reykjavik, Iceland, September 9–13,2006, Proceedings, Springer, Berlin, Heidelberg, (2006), 62–71. doi: 10.1007/11844297_7.

[10]

B. Hosseini and N. Nigam, Well-posed Bayesian inverse problems: Priors with exponential tails, SIAM/ASA J. Uncertain. Quantif., 5 (2017), 436-465. doi: 10.1137/16M1076824.

[11]

J. Kaipio and E. Somersalo, Statistical and Computational Inverse Problems volume 160 of Applied Mathematical Sciences Springer-Verlag, New York, 2005. doi: 10.1007/b138659.

[12]

C. Kraft, Some conditions for consistency and uniform consistency of statistical procedures, Univ. California Publ. Statist., 2 (1955), 125-141.

[13]

M. Lassas and S. Siltanen, Can one use total variation prior for edge-preserving Bayesian inversion?, Inverse Problems, 20 (2004), 1537-1563. doi: 10.1088/0266-5611/20/5/013.

[14]

M. LassasE. 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.

[15]

M. Ledoux and M. Talagrand, Probability in Banach Spaces: Isoperimetry and Processes volume 23 of Ergebnisse der Mathematik und ihrer Grenzgebiete (3) [Results in Mathematics and Related Areas (3)] Springer-Verlag, Berlin, 1991. doi: 10.1007/978-3-642-20212-4.

[16]

M. Markkanen, L. Roininen, J. M. J. Huttunen and S. Lasanen, Cauchy difference priors for edge-preserving Bayesian inversion with an application to X-ray tomography, 2016. arXiv: 1603. 06135v1.

[17]

J. P. Nolan, Stable Distributions — Models for Heavy Tailed Data, Birkhauser, Boston, 2017. In progress, Chapter 1 online at http://fs2.american.edu/jpnolan/www/stable/stable.html.

[18]

A. O'Hagan, Modelling with heavy tails, In Bayesian Statistics, 3 (Valencia, 1987), Oxford Sci. Publ., Oxford Univ. Press, New York, (1988), 345–359.

[19]

H. Owhadi and C. Scovel, Qualitative robustness in Bayesian inference, 2016, arXiv: 1411. 3984v3.

[20]

M. S. Pinsker, Information and Information Stability of Random Variables and Processes Holden-Day, Inc., San Francisco, Calif. -London-Amsterdam, 1964.

[21]

M. Shao and C. Nikias, Signal processing with fractional lower order moments: Stable processes and their application, Proc. IEEE, 81 (1993), 986-1010. doi: 10.1109/5.231338.

[22]

T. Steerneman, On the total variation and Hellinger distance between signed measures; an application to product measures, Proc. Amer. Math. Soc., 88 (1983), 684-688. doi: 10.1090/S0002-9939-1983-0702299-0.

[23]

A. M. Stuart, Inverse problems: A Bayesian perspective, Acta Numer., 19 (2010), 451-559. doi: 10.1017/S0962492910000061.

[24]

A. N. Tikhonov, On the solution of incorrectly put problems and the regularisation method, In Outlines Joint Sympos. Partial Differential Equations (Novosibirsk, 1963), pages 261–265. Acad. Sci. USSR Siberian Branch, Moscow, 1963.

[25]

P. TsakalidesP. Reveliotis and C. L. Nikias, Scalar quantisation of heavy-tailed signals, IEE Proc. -Vis. Image Sign., 147 (2000), 475-484. doi: 10.1049/ip-vis:20000470.

[26]

E. Tsionas, Monte Carlo inference in econometric models with symmetric stable distributions, J. Economet, 88 (1999), 365-401. doi: 10.1016/S0304-4076(98)00039-6.

show all references

References:
[1]

A. AchimP. Tsakalides and A. Bezerianos, SAR image denoising via Bayesian wavelet shrinkage based on heavy-tailed modeling, IEEE Trans. Geosci. Remote, 41 (2003), 1773-1784. doi: 10.1109/TGRS.2003.813488.

[2]

C. D. Aliprantis and K. C. Border, Infinite Dimensional Analysis: A Hitchhiker's Guide Springer, Berlin, third edition, 2006. doi: 10.1007/3-540-29587-9.

[3]

V. I. Bogachev, Differentiable Measures and the Malliavin Calculus volume 164 of Mathematical Surveys and Monographs, American Mathematical Society, Providence, RI, 2010. doi: 10.1090/surv/164.

[4]

R. Bonic, Some properties of Hilbert scales, Proc. Amer. Math. Soc., 18 (1967), 1000-1003. doi: 10.1090/S0002-9939-1967-0230115-3.

[5]

J. M. ChambersC. L. Mallows and B. W. Stuck, A method for simulating stable random variables, J. Amer. Statist. Assoc., 71 (1976), 340-344. doi: 10.1080/01621459.1976.10480344.

[6]

O. Christensen and D. T. Stoeva, p-frames in separable Banach spaces, Adv. Comput. Math., 18 (2003), 117-126. doi: 10.1023/A:1021364413257.

[7]

M. Dashti and A. M. Stuart, The Bayesian approach to inverse problems, Handbook of Uncertainty Quantification, (2016), 311-428. doi: 10.1007/978-3-319-11259-6_7-1.

[8]

M. DashtiS. Harris and A. M. Stuart, Besov priors for Bayesian inverse problems, Inverse Probl. Imaging, 6 (2012), 183-200. doi: 10.3934/ipi.2012.6.183.

[9]

N. Hansen, F. Gemperle, A. Auger and P. Koumoutsakos, When do heavy-tail distributions help?, In T. P. Runarsson, H. -G. Beyer, E. Burke, J. J. Merelo-Guervós, L. D. Whitley, and X. Yao, editors, Parallel Problem Solving from Nature — PPSN Ⅸ: 9th International Conference, Reykjavik, Iceland, September 9–13,2006, Proceedings, Springer, Berlin, Heidelberg, (2006), 62–71. doi: 10.1007/11844297_7.

[10]

B. Hosseini and N. Nigam, Well-posed Bayesian inverse problems: Priors with exponential tails, SIAM/ASA J. Uncertain. Quantif., 5 (2017), 436-465. doi: 10.1137/16M1076824.

[11]

J. Kaipio and E. Somersalo, Statistical and Computational Inverse Problems volume 160 of Applied Mathematical Sciences Springer-Verlag, New York, 2005. doi: 10.1007/b138659.

[12]

C. Kraft, Some conditions for consistency and uniform consistency of statistical procedures, Univ. California Publ. Statist., 2 (1955), 125-141.

[13]

M. Lassas and S. Siltanen, Can one use total variation prior for edge-preserving Bayesian inversion?, Inverse Problems, 20 (2004), 1537-1563. doi: 10.1088/0266-5611/20/5/013.

[14]

M. LassasE. 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.

[15]

M. Ledoux and M. Talagrand, Probability in Banach Spaces: Isoperimetry and Processes volume 23 of Ergebnisse der Mathematik und ihrer Grenzgebiete (3) [Results in Mathematics and Related Areas (3)] Springer-Verlag, Berlin, 1991. doi: 10.1007/978-3-642-20212-4.

[16]

M. Markkanen, L. Roininen, J. M. J. Huttunen and S. Lasanen, Cauchy difference priors for edge-preserving Bayesian inversion with an application to X-ray tomography, 2016. arXiv: 1603. 06135v1.

[17]

J. P. Nolan, Stable Distributions — Models for Heavy Tailed Data, Birkhauser, Boston, 2017. In progress, Chapter 1 online at http://fs2.american.edu/jpnolan/www/stable/stable.html.

[18]

A. O'Hagan, Modelling with heavy tails, In Bayesian Statistics, 3 (Valencia, 1987), Oxford Sci. Publ., Oxford Univ. Press, New York, (1988), 345–359.

[19]

H. Owhadi and C. Scovel, Qualitative robustness in Bayesian inference, 2016, arXiv: 1411. 3984v3.

[20]

M. S. Pinsker, Information and Information Stability of Random Variables and Processes Holden-Day, Inc., San Francisco, Calif. -London-Amsterdam, 1964.

[21]

M. Shao and C. Nikias, Signal processing with fractional lower order moments: Stable processes and their application, Proc. IEEE, 81 (1993), 986-1010. doi: 10.1109/5.231338.

[22]

T. Steerneman, On the total variation and Hellinger distance between signed measures; an application to product measures, Proc. Amer. Math. Soc., 88 (1983), 684-688. doi: 10.1090/S0002-9939-1983-0702299-0.

[23]

A. M. Stuart, Inverse problems: A Bayesian perspective, Acta Numer., 19 (2010), 451-559. doi: 10.1017/S0962492910000061.

[24]

A. N. Tikhonov, On the solution of incorrectly put problems and the regularisation method, In Outlines Joint Sympos. Partial Differential Equations (Novosibirsk, 1963), pages 261–265. Acad. Sci. USSR Siberian Branch, Moscow, 1963.

[25]

P. TsakalidesP. Reveliotis and C. L. Nikias, Scalar quantisation of heavy-tailed signals, IEE Proc. -Vis. Image Sign., 147 (2000), 475-484. doi: 10.1049/ip-vis:20000470.

[26]

E. Tsionas, Monte Carlo inference in econometric models with symmetric stable distributions, J. Economet, 88 (1999), 365-401. doi: 10.1016/S0304-4076(98)00039-6.

Figure 1.  Uniform angular measure on a circle projects radially to give Cauchy measure with width parameter $\gamma$ on any line at distance $\gamma$ from the centre of the circle
Figure 2.  Cauchy and Gaussian wavelet expansions in the linear spline orthonormal basis of $L^{2}([0, 1], \mathrm{d} x)$. Each horizontal stripe shows a random function $u(x) = \sum_{j = 0}^{J} \sum_{k = 0}^{2^{j} - 1} u_{j, k} 2^{j / 2} \psi(2^{j} x - k)$, where each $u_{j , k} = (j + 1)^{-2} 2^{-j}$ times a standard Cauchy or normal draw, and $\psi$ denotes the mother wavelet. The plots show $20$ i.i.d. samples with $J = 10$. Theorem 3.4 ensures a.s. convergence in $L^{2}([0, 1])$ as $J \to \infty$. To enable easy comparisons between plots, the ensemble has been translated and linearly scaled to take values $u(x) \in [0, 1]$, and the same random seed is used in each case
[1]

Nan-Jing Huang, Xian-Jun Long, Chang-Wen Zhao. Well-Posedness for vector quasi-equilibrium problems with applications. Journal of Industrial & Management Optimization, 2009, 5 (2) : 341-349. doi: 10.3934/jimo.2009.5.341

[2]

Hongjie Dong. Dissipative quasi-geostrophic equations in critical Sobolev spaces: Smoothing effect and global well-posedness. Discrete & Continuous Dynamical Systems - A, 2010, 26 (4) : 1197-1211. doi: 10.3934/dcds.2010.26.1197

[3]

M. H. Li, S. J. Li, W. Y. Zhang. Levitin-Polyak well-posedness of generalized vector quasi-equilibrium problems. Journal of Industrial & Management Optimization, 2009, 5 (4) : 683-696. doi: 10.3934/jimo.2009.5.683

[4]

Byeongchan Lee, Jonghun Yoon, Yang Woo Shin, Ganguk Hwang. Tail asymptotics of fluid queues in a distributed server system fed by a heavy-tailed ON-OFF flow. Journal of Industrial & Management Optimization, 2016, 12 (2) : 637-652. doi: 10.3934/jimo.2016.12.637

[5]

Yuebao Wang, Qingwu Gao, Kaiyong Wang, Xijun Liu. Random time ruin probability for the renewal risk model with heavy-tailed claims. Journal of Industrial & Management Optimization, 2009, 5 (4) : 719-736. doi: 10.3934/jimo.2009.5.719

[6]

Yang Yang, Kam C. Yuen, Jun-Feng Liu. Asymptotics for ruin probabilities in Lévy-driven risk models with heavy-tailed claims. Journal of Industrial & Management Optimization, 2018, 14 (1) : 231-247. doi: 10.3934/jimo.2017044

[7]

Timur Akhunov. Local well-posedness of quasi-linear systems generalizing KdV. Communications on Pure & Applied Analysis, 2013, 12 (2) : 899-921. doi: 10.3934/cpaa.2013.12.899

[8]

Qunyi Bie, Qiru Wang, Zheng-An Yao. On the well-posedness of the inviscid Boussinesq equations in the Besov-Morrey spaces. Kinetic & Related Models, 2015, 8 (3) : 395-411. doi: 10.3934/krm.2015.8.395

[9]

Masoumeh Dashti, Stephen Harris, Andrew Stuart. Besov priors for Bayesian inverse problems. Inverse Problems & Imaging, 2012, 6 (2) : 183-200. doi: 10.3934/ipi.2012.6.183

[10]

Kenji Kimura, Yeong-Cheng Liou, Soon-Yi Wu, Jen-Chih Yao. Well-posedness for parametric vector equilibrium problems with applications. Journal of Industrial & Management Optimization, 2008, 4 (2) : 313-327. doi: 10.3934/jimo.2008.4.313

[11]

Nigel Boston, Jing Hao. The weight distribution of quasi-quadratic residue codes. Advances in Mathematics of Communications, 2018, 12 (2) : 363-385. doi: 10.3934/amc.2018023

[12]

G. Fonseca, G. Rodríguez-Blanco, W. Sandoval. Well-posedness and ill-posedness results for the regularized Benjamin-Ono equation in weighted Sobolev spaces. Communications on Pure & Applied Analysis, 2015, 14 (4) : 1327-1341. doi: 10.3934/cpaa.2015.14.1327

[13]

Hongjie Dong, Dapeng Du. Global well-posedness and a decay estimate for the critical dissipative quasi-geostrophic equation in the whole space. Discrete & Continuous Dynamical Systems - A, 2008, 21 (4) : 1095-1101. doi: 10.3934/dcds.2008.21.1095

[14]

Guillaume Bal, Ian Langmore, Youssef Marzouk. Bayesian inverse problems with Monte Carlo forward models. Inverse Problems & Imaging, 2013, 7 (1) : 81-105. doi: 10.3934/ipi.2013.7.81

[15]

Nikolaos Bournaveas. Local well-posedness for a nonlinear dirac equation in spaces of almost critical dimension. Discrete & Continuous Dynamical Systems - A, 2008, 20 (3) : 605-616. doi: 10.3934/dcds.2008.20.605

[16]

Stefan Meyer, Mathias Wilke. Global well-posedness and exponential stability for Kuznetsov's equation in $L_p$-spaces. Evolution Equations & Control Theory, 2013, 2 (2) : 365-378. doi: 10.3934/eect.2013.2.365

[17]

Zhichun Zhai. Well-posedness for two types of generalized Keller-Segel system of chemotaxis in critical Besov spaces. Communications on Pure & Applied Analysis, 2011, 10 (1) : 287-308. doi: 10.3934/cpaa.2011.10.287

[18]

Ming Wang. Sharp global well-posedness of the BBM equation in $L^p$ type Sobolev spaces. Discrete & Continuous Dynamical Systems - A, 2016, 36 (10) : 5763-5788. doi: 10.3934/dcds.2016053

[19]

Kai Yan, Zhaoyang Yin. Well-posedness for a modified two-component Camassa-Holm system in critical spaces. Discrete & Continuous Dynamical Systems - A, 2013, 33 (4) : 1699-1712. doi: 10.3934/dcds.2013.33.1699

[20]

Allaberen Ashyralyev. Well-posedness of the modified Crank-Nicholson difference schemes in Bochner spaces. Discrete & Continuous Dynamical Systems - B, 2007, 7 (1) : 29-51. doi: 10.3934/dcdsb.2007.7.29

2016 Impact Factor: 1.094

Metrics

  • PDF downloads (1)
  • HTML views (1)
  • Cited by (2)

Other articles
by authors

[Back to Top]