Abstract
One way to quantify the uncertainty in Bayesian inverse problems arising in the engineering domain is to generate samples from the posterior distribution using Markov chain Monte Carlo (MCMC) algorithms. The basic MCMC methods tend to explore the parameter space slowly, which makes them inefficient for practical problems. On the other hand, enhanced MCMC approaches, like Hamiltonian Monte Carlo (HMC), require the gradients from the physical problem simulator, which are often not available. In this case, a feasible option is to use the gradient approximations provided by the surrogate (proxy) models built on the simulator output. In this paper, we consider proxy-aided HMC employing the Gaussian process (kriging) emulator. We overview in detail the different aspects of kriging proxies, the underlying principles of the HMC sampler and its interaction with the proxy model. The proxy-aided HMC algorithm is thoroughly tested in different settings, and applied to three case studies—one toy problem, and two synthetic reservoir simulation models. We address the question of how the sampler performance is affected by the increase of the problem dimension, the use of the gradients in proxy training, the use of proxy-for-the-data and the different approaches to the design points selection. It turns out that applying the proxy model with HMC sampler may be beneficial for relatively small physical models, with around 20 unknown parameters. Such a sampler is shown to outperform both the basic Random Walk Metropolis algorithm, and the HMC algorithm fed by the exact simulator gradients.
Article PDF
Similar content being viewed by others
References
Punq-S3 model. Available from: https://www.imperial.ac.uk/earth-science/research/research-groups/perm/standard-models/
Abdollahzadeh A, Reynolds A, Christie M, Corne D, Davies B, and Williams G, (2011) Bayesian optimization algorithm applied to uncertainty quantification, in SPE-143290-MS
Beck, J., Guillas, S.: Sequential design with mutual information for computer experiments (MICE): emulation of a tsunami model. SIAM/ASA J Uncertain Quantif. 4(1), 739–766 (2016)
Betancourt M, Byrne S, and M. Girolami, (2015) Optimizing the integrator step size for Hamiltonian Monte Carlo. arXiv
Bilionis, I., Zabaras, N.: Multi-output local Gaussian process regression: applications to uncertainty quantification. J Comput Phys. 231, 5718–5746 (2012)
Bou-Rabee N and Sanz-Serna JM, Geometric integrators and the Hamiltonian Monte Carlo method. Acta Numerica, 2018: p. 1–92
Bui-Thanh, T., Ghattas, O., Higdon, D.: Adaptive Hessian-based nonstationary Gaussian process response surface method for probability density approximation with application to Bayesian solution of large-scale inverse problems. SIAM J Sci Comput. 34, A2837–AA287 (2011)
Busby, D., Da Veiga, S., Touzani, S.: A workflow for decision making under uncertainty. Comput Geosci. 18, 519–533 (2014)
Christen, J.A., Sanso, B.: Advances in the sequential design of computer experiments based on active learning. Commun Statistics - Theory Methods. 40, 4467–4483 (2011)
Conrad, P., Marzouk, Y., Pillai, N., Smith, A.: Accelerating asymptotically exact MCMC for computationally intensive models via local approximations. J Am Stat Assoc. 111, 1591–1607 (2015)
Eriksson, D., K. Dong, E.H. Lee, D. Bindel, and A.G. Wilson. Scaling Gaussian process regression with derivatives. in Neural information processing systems 2018
Floris, F.J.T., Bush, M.D., Cuypers, M., Roggero, F., Syversveen, A.-R.: Methods for quantifying the uncertainty of production forecasts: a comparative study. Pet Geosci. 7, S87–S96 (2001)
Geyer, C.J.: Practical Markov chain Monte Carlo. Stat Sci. 7(4), 473–483 (1992)
Ginsbourger, D., Sequential design of computer experiments. Wiley StatsRef, 2017
Girolami, M., Calderhead, B.: Riemann manifold Langevin and Hamiltonian Monte Carlo methods. J R Stat Soc. 73, 123–159 (2011)
Gramacy, R.B., Lee, H.K.H.: Bayesian treed Gaussian process models with an application to computer modeling. J Am Stat Assoc. 103, 1119–1130 (2008)
Gramacy, R.B. and Lee, H.K.H, Cases for the nugget in modeling computer experiments. arXiv paper, 2010
Hajizadeh, Y., Ants can do history matching, in SPE-141137-STU. 2010
He, J., J. Xie, X.-H. Wen, and W. Chen, Improved proxy for history matching using proxy-for-data approach and reduced order modeling, in SPE-174055-MS. 2015
Hoffman, M.D., Gelman, A.: The no-U-turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo. J Mach Learn Res. 15, 1351–1381 (2014)
Jones, D.R.: A taxonomy of global optimization methods based on response surfaces. J Glob Optim. 21, 345–383 (2001)
Kaipio, J. and E. Somersalo, Statistical and computational inverse problems. 2005: Springer
Korattikara, A., Y. Chen, and M. Welling. Austerity in MCMC land: cutting the Metropolis-Hastings budget. in 31st International Conference on Machine Learning. 2014
Lan, S., Bui-Thanh, T., Christie, M., Girolami, M.: Emulation of higher-order tensors in manifold Monte Carlo methods for Bayesian inverse problems. J Comput Phys. 308, 81–101 (2016)
Lan, S., Stathopoulos, V., Shahbaba, B., Girolami, M.: Markov chain Monte Carlo from Lagrangian dynamics. J Comput Graph Stat. 24(2), 357–378 (2015)
Landa, J. and G. Baris, A methodology for history matching and the assessment of uncertainties associated with flow prediction, in SPE ATCE. 2003
Liu, J.S., Monte-Carlo strategies in scientific computing. Springer Series in Statistics. 2004: Springer; 1st edition
Ma, Y., E.B. Fox, T. Chen, and L. Wu, Irreversible samplers from jump and continuous Markov processes. arXiv, 2018
Neal, R.M., MCMC using Hamiltonian dynamics, in Handbook of Markov Chain Monte Carlo, S. Brooks, et al., Editors. 2011, Chapman & Hall / CRC Press
Oakley, J., Bayesian uncertainty analysis for complex computer codes. 1999, University of Sheffield
Oliver, D.S., A.C. Reynolds, and N. Liu, Inverse theory for petroleum reservoir characterization and history matching. 2008: Cambridge University Press
Osborne, M.A., R. Garnett, and S.J. Roberts. Gaussian processes for global optimization. in 3rd International conference on learning and intelligent optimization. 2009
Rasmussen, C.: Gaussian processes to speed up hybrid Monte Carlo for expensive Bayesian integrals. Bayesian Statistics. 7, 651–659 (2003)
Robert, C. and G. Casella, Monte Carlo Statistical Methods. 2 ed. 2004: Springer Science & Business Media
Sacks, J., Welch, W.J., Mitchell, T.J., Wynn, H.P.: Design and analysis of computer experiments. Stat Sci. 4(4), 409–423 (1989)
Solak, E., R. Murray-Smith, W.E. Leithead, D.J. Leith, and C.E. Rasmussen. Derivative observations in Gaussian process models of dynamic systems. in Neural Information Processing Systems. 2002
Stein, M.L.: Interpolation of spatial data: some theory for kriging. Springer, New York (1999)
Tarantola, A., Inverse problem theory and methods for model parameter estimation. 2005
Welling, M. and Y. Teh. Bayesian learning via stochastic gradient Langevin dynamics. in 28th International Conference on Machine Learning. 2011
Zubarev, D.I., Pros and cons of applying proxy-models as a substitute for full reservoir simulations, in SPE-124815-MS. 2009
Acknowledgements
This work would not have been possible without sponsorship of the EQUIP (Enabling Uncertainty Quantification for Inverse Problems) grant from the Engineering and Physical Sciences Research Council (EPSRC) EP/K034154/1. We also thank Schlumberger for providing the Eclipse simulator to Heriot-Watt University, and acknowledge the use of the open source software libraries OpenMPI, BLAS, LAPACK, ALGLIB.
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
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Fursov, I., Christie, M. & Lord, G. Applying kriging proxies for Markov chain Monte Carlo in reservoir simulation. Comput Geosci 24, 1725–1746 (2020). https://doi.org/10.1007/s10596-020-09968-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10596-020-09968-z