Abstract
We introduce a new Markov chain Monte Carlo (MCMC) sampler for infinite-dimensional inverse problems. Our new sampler is based on the affine invariant ensemble sampler, which uses interacting walkers to adapt to the covariance structure of the target distribution. We extend this ensemble sampler for the first time to infinite-dimensional function spaces, yielding a highly efficient gradient-free MCMC algorithm. Because our new ensemble sampler does not require gradients or posterior covariance estimates, it is simple to implement and broadly applicable.
Similar content being viewed by others
References
Akeret, J., Seehars, S., Amara, A., Refregier, A., Csillaghy, A.: Cosmohammer: cosmological parameter estimation with the MCMC hammer. Astron. Comput. 2, 27–39 (2013)
Beskos, A., Girolami, M., Lan, S., Farrell, P.E., Stuart, A.M.: Geometric MCMC for infinite-dimensional inverse problems. J. Comput. Phys. 335, 327–351 (2017)
Beskos, A., Jasra, A., Law, K., Marzouk, Y., Zhou, Y.: Multilevel sequential Monte Carlo with dimension-independent likelihood-informed proposals. SIAM/ASA J. Uncertain. Quantif. 6(2), 762–786 (2018)
Beskos, A., Roberts, G., Stuart, A., Voss, J.: MCMC methods for diffusion bridges. Stoch. Dyn. 8(03), 319–350 (2008)
Chen, Y., Keyes, D., Law, K.J., Ltaief, H.: Accelerated dimension-independent adaptive Metropolis. SIAM J. Sci. Comput. 38(5), S539–S565 (2016)
Cotter, S.L., Roberts, G.O., Stuart, A.M., White, D.: MCMC methods for functions: modifying old algorithms to make them faster. Stat. Sci. **424–446 (2013)
Coullon, J., Pokern, Y.: MCMC for a hyperbolic Bayesian inverse problem in traffic flow modelling (2020). arXiv:2001.02013
Cui, T., Law, K.J., Marzouk, Y.M.: Dimension-independent likelihood-informed MCMC. J. Comput. Phys. 304, 109–137 (2016)
Cui, T., Martin, J., Marzouk, Y.M., Solonen, A., Spantini, A.: Likelihood-informed dimension reduction for nonlinear inverse problems. Inverse Probl. 30(11), 114015 (2014)
Dunlop, M.M., Stuart, A.M.: The Bayesian formulation of EIT: analysis and algorithms. Inverse Probl. Imaging 10(4), 1007–1036 (2016)
Foreman-Mackey, D., Hogg, D.W., Lang, D., Goodman, J.: emcee: the MCMC hammer. Publ. Astron. Soc. Pac. 125(925), 306 (2013)
Goodman, J., Weare, J.: Ensemble samplers with affine invariance. Commun. Appl. Math. Comput. Sci. 5(1), 65–80 (2010)
Hairer, M., Stuart, A.M., Vollmer, S.J., et al.: Spectral gaps for a metropolis-hastings algorithm in infinite dimensions. Ann. Appl. Probab. 24(6), 2455–2490 (2014)
Hu, Z., Yao, Z., Li, J.: On an adaptive preconditioned Crank–Nicolson MCMC algorithm for infinite dimensional Bayesian inference. J. Comput. Phys. 332, 492–503 (2017)
Iglesias, M.A., Lin, K., Stuart, A.M.: Well-posed Bayesian geometric inverse problems arising in subsurface flow. Inverse Probl. 30(11), 114001 (2014). https://doi.org/10.1088/0266-5611/30/11/114001
Kantas, N., Beskos, A., Jasra, A.: Sequential Monte Carlo methods for high-dimensional inverse problems: a case study for the Navier–Stokes equations. SIAM/ASA J. Uncertain. Quantif. 2(1), 464–489 (2014)
Law, K.J.: Proposals which speed up function-space MCMC. J. Comput. Appl. Math. 262, 127–138 (2014)
Metropolis, N., Rosenbluth, A.W., Rosenbluth, M.N., Teller, A.H., Teller, E.: Equation of state calculations by fast computing machines. J. Chem. Phys. 21(6), 1087–1092 (1953)
Roberts, G.O., Rosenthal, J.S.: Examples of adaptive MCMC. J. Comput. Graph. Stat. 18(2), 349–367 (2009)
Rudolf, D., Sprungk, B.: On a generalization of the preconditioned Crank–Nicolson Metropolis algorithm. Found. Comput. Math. 18(2), 309–343 (2018)
Sokal, A.: Monte Carlo methods in statistical mechanics: Foundations and new algorithms. In: Functional Integration, pp. 131–192. Springer (1997)
Stuart, A.M.: Inverse problems: a Bayesian perspective. Acta Numer. 19, 451 (2010)
Zhou, Q., Hu, Z., Yao, Z., Li, J.: A hybrid adaptive MCMC algorithm in function spaces. SIAM/ASA J. Uncertain. Quantifi. 5(1), 621–639 (2017)
Acknowledgements
We thank Christopher Nemeth, Gideon Simpson, and Jonathan Weare for offering useful critiques. JC was supported by EPSRC grant EP/S00159X/1. RJW was supported by the National Science Foundation award DMS-1646339 and by New York University’s Dean’s Dissertation Fellowship. The High End Computing facility at Lancaster University provided computing resources.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Coullon, J., Webber, R.J. Ensemble sampler for infinite-dimensional inverse problems . Stat Comput 31, 28 (2021). https://doi.org/10.1007/s11222-021-10004-y
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11222-021-10004-y
Keywords
- Bayesian inverse problems
- Markov chain Monte Carlo
- Infinite-dimensional inverse problems
- Dimensionality reduction