Emulation-accelerated Hamiltonian Monte Carlo algorithms for parameter estimation and uncertainty quantification in differential equation models. (English) Zbl 1477.62014
Summary: We propose to accelerate Hamiltonian and Lagrangian Monte Carlo algorithms by coupling them with Gaussian processes for emulation of the log unnormalised posterior distribution. We provide proofs of detailed balance with respect to the exact posterior distribution for these algorithms, and validate the correctness of the samplers’ implementation by Geweke consistency tests. We implement these algorithms in a delayed acceptance (DA) framework, and investigate whether the DA scheme can offer computational gains over the standard algorithms. A comparative evaluation study is carried out to assess the performance of the methods on a series of models described by differential equations, including a real-world application of a 1D fluid-dynamics model of the pulmonary blood circulation. The aim is to identify the algorithm which gives the best trade-off between accuracy and computational efficiency, to be used in nonlinear DE models, which are computationally onerous due to repeated numerical integrations in a Bayesian analysis. Results showed no advantage of the DA scheme over the standard algorithms with respect to several efficiency measures based on the effective sample size for most methods and DE models considered. These gradient-driven algorithms register a high acceptance rate, thus the number of expensive forward model evaluations is not significantly reduced by the first emulator-based stage of DA. Additionally, the Lagrangian Dynamical Monte Carlo and Riemann Manifold Hamiltonian Monte Carlo tended to register the highest efficiency (in terms of effective sample size normalised by the number of forward model evaluations), followed by the Hamiltonian Monte Carlo, and the No U-turn sampler tended to be the least efficient.
Keywords:
parameter estimation; uncertainty quantification; MCMC; emulation; Gaussian processes; differential equationsReferences:
| [1] | Álvarez, M., Luengo, D., Titsias, M., Lawrence, N.D.: Efficient multioutput Gaussian processes through variational inducing kernels. In: Teh, Y.W., Titterington, M. (eds) Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, PMLR, Chia Laguna Resort, Sardinia, Italy, Proceedings of Machine Learning Research, vol. 9, pp. 25-32 (2010) |
| [2] | Banterle, M.; Grazian, C.; Lee, A.; Robert, CP, Accelerating Metropolis-Hastings algorithms by delayed acceptance, Found. Data Sci., 1, 2, 103-128 (2019) |
| [3] | Bastos, L.S., O’Hagan, A.: Diagnostics for Gaussian process emulators. Technometrics 51(4), 425-438 (2009) |
| [4] | Betancourt M (2015) The fundamental incompatibility of scalable Hamiltonian Monte Carlo and Naive data subsampling. In: Bach F, Blei D (eds) Proceedings of the 32nd International Conference on Machine Learning, PMLR, Lille, France, Proceedings of Machine Learning Research, vol. 37, pp 533-540 |
| [5] | Bliznyuk, N.; Ruppert, D.; Shoemaker, C.; Regis, R.; Wild, S.; Mugunthan, P., Bayesian calibration and uncertainty analysis for computationally expensive models using optimization and radial basis function approximation, J. Comput. Graph. Stat., 17, 270-294 (2008) |
| [6] | Bowman, AW; Azzalini, A., Applied Smoothing Techniques for Data Analysis (1997), New York: Oxford University Press Inc, New York · Zbl 0889.62027 |
| [7] | Bratley, P., Fox, B.: Algorithm 659: implementing Sobol’s quasirandom sequence generator. ACM Trans. Math. Softw. 14(1), 88-100 (1988) · Zbl 0642.65003 |
| [8] | Brooks, S.; Gelman, A., General methods for monitoring convergence of iterative simulations, J. Comput. Graph. Stat., 7, 4, 434-455 (1998) |
| [9] | Brooks, S., Gelman, A., Jones, G.L., Meng, X.L. (eds.): Handbook of Markov Chain Monte Carlo. Handbooks of Modern Statistical Methods, Chapman and Hall (2011) · Zbl 1218.65001 |
| [10] | Broyden, CG, Quasi-Newton Methods (1972), London: Academic Press, London · Zbl 0277.65031 |
| [11] | Brüggemeier, B., Schusterreiter, C., Pavlou, H., Jenkins, N., Lynch, S., Bianchi, A., Cai, X.: Improving the utility of drosophila melanogaster for neurodegenerative disease research by modelling courtship behaviour patterns. In: Report Summarising the Outcomes from the UK NC3R’s and POEM’s Meeting (2014) |
| [12] | Bui-Thanh, T.; Girolami, M., Solving large-scale PDE-constrained Bayesian inverse problems with Riemann Manifold Hamiltonian Monte Carlo, Inverse Prob., 30, 11, 114014 (2014) · Zbl 1306.65269 |
| [13] | Calderhead, B.: Differential Geometric MCMC Methods and Applications. PhD Thesis, University of Glasgow (2012) |
| [14] | Campbell, D.; Steele, RJ, Smooth functional tempering for nonlinear differential equation models, Stat. Comput., 22, 429-443 (2012) · Zbl 1322.62011 |
| [15] | Chen, T., Fox, E., Guestrin, C.: Stochastic Gradient Hamiltonian Monte Carlo. In: 31st International Conference on Machine Learning, ICML vol. 5 (2014) |
| [16] | Christen, J.; Fox, C., Markov Chain Monte Carlo using an approximation, J. Comput. Graph. Stat., 14, 4, 795-810 (2005) |
| [17] | Conrad, PR; Marzouk, YM; Pillai, NS; Smith, A., Accelerating asymptotically exact MCMC for computationally intensive models via local approximations, J. Am. Stat. Assoc., 111, 516, 1591-1607 (2016) |
| [18] | Conrad, PR; Davis, AD; Marzouk, YM; Pillai, NS; Smith, A., Parallel local approximation MCMC for expensive models, SIAM/ASA J. Uncertain. Quantif., 6, 1, 339-373 (2018) · Zbl 1398.65017 |
| [19] | Conti, S., O’Hagan, A.: Bayesian emulation of complex multi-output and dynamic computer models. J. Stat. Plan. Inference 140(3), 640-651 (2010) · Zbl 1177.62033 |
| [20] | Conti, S., O’Hagan, A.: Bayesian emulation of complex multi-output and dynamic computer models. J. Stat. Plan. Inference 140(3), 640-651 (2010) · Zbl 1177.62033 |
| [21] | Conti, S., Gosling, J.P., Oakley, J.E., O’Hagan, A.: Gaussian process emulation of dynamic computer codes. Biometrika 96(3), 663-676 (2009) · Zbl 1437.62015 |
| [22] | Costabal, FS; Matsuno, K.; Yao, J.; Perdikaris, P.; Kuhl, E., Machine learning in drug development: characterizing the effect of 30 drugs on the QT interval using Gaussian process regression, sensitivity analysis, and uncertainty quantification, Comput. Methods Appl. Mech. Eng., 348, 313-333 (2019) · Zbl 1440.62371 |
| [23] | Cui, T., Fox, C., O’Sullivan, M.: Bayesian calibration of a large-scale geothermal reservoir model by a new adaptive delayed acceptance Metropolis Hastings algorithm. Water Resour. Res. 47(10), W10521 (2011) |
| [24] | Davies, V.; Noé, U.; Lazarus, A.; Gao, H.; Macdonald, B.; Berry, C.; Luo, X.; Husmeier, D., Fast parameter inference in a biomechanical model of the left ventricle by using statistical emulation, J. R. Stat. Soc. Ser. C: Appl. Stat., 68, 5, 1555-1576 (2019) |
| [25] | Dietzel, A.; Reichert, P., Bayesian inference of a lake water quality model by emulating its posterior density, Water Resour. Res., 50, 10, 7626-7647 (2014) |
| [26] | Drovandi, CC; Moores, MT; Boys, RJ, Accelerating pseudo-marginal MCMC using Gaussian processes, Comput. Stat. Data Anal., 118, 1-17 (2018) · Zbl 1469.62057 |
| [27] | Fielding, M.; Nott, D.; Liong, SY, Efficient MCMC schemes for computationally expensive posterior distributions, Technometrics, 53, 16-28 (2011) |
| [28] | Fitzhugh, R., Impulses and physiological states in theoretical models of nerve membrane, Biophys. J ., 1, 6, 445-466 (1961) |
| [29] | Geweke, J., Getting it right: joint distribution tests of posterior simulators, J. Am. Stat. Assoc., 99, 467, 799-804 (2004) · Zbl 1117.62344 |
| [30] | Girolami, M.; Calderhead, B., Riemann Manifold Langevin and Hamiltonian Monte Carlo methods, J. R. Stat. Soc. Ser. B (Stat. Methodol.), 73, 2, 123-214 (2011) · Zbl 1411.62071 |
| [31] | Golightly, A.; Henderson, DA; Sherlock, C., Delayed acceptance particle MCMC for exact inference in stochastic kinetic models, Stat. Comput., 25, 5, 1039-1055 (2015) · Zbl 1332.60117 |
| [32] | Gong, W.; Duan, Q., An adaptive surrogate modeling-based sampling strategy for parameter optimization and distribution estimation (ASMO-PODE), Environ. Modell. Softw., 95, 61-75 (2017) |
| [33] | Göktepe, S.; Kuhl, E., Computational modeling of cardiac electrophysiology: a novel finite element approach, Int. J. Numer. Methods Eng., 79, 2, 156-178 (2009) · Zbl 1171.92310 |
| [34] | Haario, H.; Saksman, E.; Tamminen, J., An adaptive Metropolis algorithm, Bernoulli, 7, 2, 223-242 (2001) · Zbl 0989.65004 |
| [35] | Hensman, J., Matthews, A.G., Filippone, M., Ghahramani, Z. MCMC for variationally sparse Gaussian processes. In: Cortes, C., Lawrence, N., Lee, D., Sugiyama, M., Garnett, R. (eds) Advances in Neural Information Processing Systems. Curran Associates Inc., vol. 28 (2015) |
| [36] | Higdon, D., Reese, C., Moulton, J., Vrugt, J., Fox, C.: Posterior exploration for computationally intensive forward models, pp. 401-418 (2011) · Zbl 1229.65016 |
| [37] | Hoffman, M.; Gelman, A., The no-U-turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo, J. Mach. Learn. Res., 15, 1, 1593-1623 (2014) · Zbl 1319.60150 |
| [38] | Jones, D.; Schonlau, M.; Welch, W., Efficient global optimization of expensive black-box functions, J. Glob. Optim., 13, 455-492 (1998) · Zbl 0917.90270 |
| [39] | Kass, R.; Carlin, B.; Gelman, A.; Neal, R., Markov Chain Monte Carlo in practice: a roundtable discussion, Am. Stat., 52, 2, 93-100 (1998) |
| [40] | Kennedy, M.C., O’Hagan, A.: Bayesian calibration of computer models. J. R. Stat. Soc. Ser. B (Stat. Methodol.) 63(3), 425-464 (2001) · Zbl 1007.62021 |
| [41] | Kim, E.K., Choi, E.J.: Pathological roles of MAPK signaling pathways in human diseases. Biochim. Biophysi. Acta (BBA) Mol. Basis Disease 1802(4), 396-405 (2010) |
| [42] | Kramer, A.; Calderhead, B.; Radde, N., Hamiltonian Monte Carlo methods for efficient parameter estimation in steady state dynamical systems, BMC Bioinform., 15, 1, 253 (2014) |
| [43] | Laine, M.: MCMC Toolbox for Matlab. http://helios.fmi.fi/ lainema/dram/ (2007) |
| [44] | Lan, S.; Stathopoulos, V.; Shahbaba, B.; Girolami, M., Markov Chain Monte Carlo from Lagrangian dynamics, J. Comput. Graph. Stat., 24, 2, 357-378 (2015) |
| [45] | 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) · Zbl 1352.65010 |
| [46] | Lawrence, ND; Girolami, M.; Rattray, M.; Sanguinetti, G., Learning and Inference in Computational Systems Biology (2010), Cambridge: MIT Press, Cambridge · Zbl 1196.92018 |
| [47] | Lax, P.; Wendroff, B., Systems of conservation laws, Commun. Pure Appl. Math., 13, 2, 217-237 (1960) · Zbl 0152.44802 |
| [48] | Livingstone, S.; Betancourt, M.; Byrne, S.; Girolami, M., On the geometric ergodicity of Hamiltonian Monte Carlo, Bernoulli, 25, 4, 3109-3138 (2019) · Zbl 1428.62375 |
| [49] | Lê, M.; Delingette, H.; Kalpathy-Cramer, J.; Gerstner, ER; Batchelor, T.; Unkelbach, J.; Ayache, N., MRI based Bayesian personalization of a tumor growth model, IEEE Trans. Med. Imaging, 35, 10, 2329-2339 (2016) |
| [50] | Martin, G., Cell signaling and cancer, Cancer Cell, 4, 3, 167-174 (2004) |
| [51] | McKay, MD; Beckman, RJ; Conover, WJ, Comparison of three methods for selecting values of input variables in the analysis of output from a computer code, Technometrics, 21, 2, 239-245 (1979) · Zbl 0415.62011 |
| [52] | Mirams, GR; Pathmanathan, P.; Gray, RA; Challenor, P.; Clayton, RH, Uncertainty and variability in computational and mathematical models of cardiac physiology, J. Physiol., 594, 23, 6833-6847 (2016) |
| [53] | Mockus, J.; Tiesis, V.; Zilinskas, A., The application of Bayesian methods for seeking the extremum, Towards Glob. Optim., 2, 117-129 (1978) · Zbl 0394.90090 |
| [54] | Moreno-Muñoz, P., Artés, A., Álvarez, M.: Heterogeneous multi-output Gaussian process prediction. In: Bengio, S., Wallach, H., Larochelle, H., Grauman, K., Cesa-Bianchi, N., Garnett, R. (eds) Advances in Neural Information Processing Systems 31, pp. 6711-6720. Curran Associates, Inc. (2018) |
| [55] | Murphy, KP, Machine Learning: A Probabilistic Approach (2012), Cambridge: MIT Press, Cambridge · Zbl 1295.68003 |
| [56] | Murray, I., Graham, M.: Pseudo-marginal slice sampling. In: Gretton, A., Robert, C.C. (eds) Proceedings of the 19th International Conference on Artificial Intelligence and Statistics, Cadiz, Spain, JMLR: W&CP, vol. 51, pp. 911-919 (2016) |
| [57] | Nagumo, J.; Arimoto, S.; Yoshizawa, S., An active pulse transmission line simulating nerve axon, Proc. IRE, 50, 10, 2061-2070 (1962) |
| [58] | Neal, R., MCMC using Hamiltonian dynamics, Handb. Markov Chain Monte Carlo, 2, 113-162 (2011) · Zbl 1229.65018 |
| [59] | Olufsen, M.; Peskin, CS; Kim, W.; Pedersen, EM; Nadim, A.; Larsen, J., Numerical simulation and experimental validation of blood flow in arteries with structured-tree outflow conditions, Ann. Biomed. Eng., 28, 1281-1299 (2000) |
| [60] | Paun, L.M., Husmeier, D.: Markov chain Monte Carlo with Gaussian processes for fast parameter estimation and uncertainty quantification in a 1D fluid-dynamics model of the pulmonary circulation. Int. J. Numer. Methods Biomed. Eng. p. e3421 (2020) |
| [61] | Paun, L.M., Colebank, M., Umar Qureshi, M., Olufsen, M., Hill, N., Husmeier, D.: MCMC with Delayed Acceptance using a Surrogate Model with an Application to Cardiovascular Fluid Dynamics. In: Proceedings of the International Conference on Statistics: Theory and Applications (ICSTA’19) (2019) |
| [62] | Peirlinck, M.; Sahli Costabal, F.; Sack, KL; Choy, JS; Kassab, GS; Guccione, JM; De Beule, M.; Segers, P.; Kuhl, E., Using machine learning to characterize heart failure across the scales, Biomech. Model. Mechanobiol., 18, 6, 1987-2001 (2019) |
| [63] | Quiroz, M.; Tran, MN; Villani, M.; Kohn, R., Speeding up MCMC by delayed acceptance and data subsampling, J. Comput. Graph. Stat., 27, 1, 12-22 (2018) · Zbl 1420.62121 |
| [64] | Qureshi, M., Haider, M., Chesler, N., Olufsen, M.: Simulating effects of hypoxia on pulmonary haemodynamics in mice. In: Proceedings of the 5th International Conference on Computational and Mathematical Biomedical Engineering (CMBE 2017), vol 1, pp. 271-274. Zeta Computational Resources Ltd. (2017) |
| [65] | Qureshi, M.; Colebank, M.; Paun, L.; Chesler, N.; Haider, M.; Hill, N.; Husmeier, D.; Olufsen, M., A computational study of pulmonary hemodynamics in healthy and hypoxic mice, Biomech. Modell. Mechanobiol., 18, 1, 219-243 (2018) |
| [66] | Ramsay, JO; Hooker, G.; Campbell, D.; Cao, J., Parameter estimation for differential equations: a generalized smoothing approach, J. R. Stat. Soc. B, 69, 5, 741-796 (2007) · Zbl 07555374 |
| [67] | Rasmussen, C., Gaussian processes to speed up hybrid Monte Carlo for expensive Bayesian integrals, Bayesian Stat., 7, 7, 651-659 (2003) |
| [68] | Rasmussen, C.; Williams, C., Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning) (2005), Cambridge: The MIT Press, Cambridge |
| [69] | Schiavazzi, D.; Arbia, G.; Baker, C.; Hlavacek, A.; Hsia, T.; Marsden, A.; Vignon-Clementel, I., Uncertainty quantification in virtual surgery hemodynamics predictions for single ventricle palliation, Int. J. Numer. Methods Biomed. Eng., 32, 3, e02737 (2016) |
| [70] | Sengupta, B.; Friston, K.; Penny, W., Efficient gradient computation for dynamical models, Neuroimage, 98, 521-527 (2014) |
| [71] | Sengupta, B.; Friston, K.; Penny, W., Gradient-based MCMC samplers for dynamic causal modelling, Neuroimage, 125, 1107-1118 (2016) |
| [72] | Shahriari, B.; Swersky, K.; Wang, Z.; Adams, RP; de Freitas, N., Taking the human out of the loop: a review of Bayesian optimization, Proc. IEEE, 104, 148-175 (2016) |
| [73] | Sherlock, C.; Golightly, A.; Henderson, D., Adaptive, delayed-acceptance MCMC for targets with expensive likelihoods, J. Comput. Graph. Stat., 26, 2, 434-444 (2017) |
| [74] | Titsias M (2009) Variational learning of inducing variables in sparse Gaussian processes. In: van Dyk, D., Welling, M. (eds) Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics, PMLR, Hilton Clearwater Beach Resort, Clearwater Beach, Florida USA, Proceedings of Machine Learning Research, vol. 5, pp. 567-574 |
| [75] | Turner, B.; Sederberg, P.; Brown, S.; Steyvers, M., A method for efficiently sampling from distributions with correlated dimensions, Psychol. Methods, 18, 3, 368-384 (2013) |
| [76] | Vanhatalo, J.; Riihimäki, J.; Hartikainen, J.; Jylänki, P.; Tolvanen, V.; Vehtari, A., GPstuff: Bayesian modeling with Gaussian processes, J. Mach. Learn. Res., 14, 1, 1175-1179 (2013) · Zbl 1320.62010 |
| [77] | Wang, Z., Mohamed, S., de Freitas, N.: Adaptive Hamiltonian and Riemann Manifold Monte Carlo samplers. In: Proceedings of the 30th International Conference on International Conference on Machine Learning—Vol. 28, JMLR.org, ICML’13, pp. III-1462-III-1470 (2013) |
| [78] | Wilkinson, DJ, Bayesian methods in bioinformatics and computational systems biology, Brief. Bioinform., 8, 2, 109-116 (2007) |
| [79] | Wilkinson, R.: Accelerating ABC methods using Gaussian processes. In: Kaski, S., Corander, J. (eds) Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, PMLR, Reykjavik, Iceland, Proceedings of Machine Learning Research, vol. 33, pp. 1015-1023 (2014) |
| [80] | Wu, K.; Li, J., A surrogate accelerated multicanonical Monte Carlo method for uncertainty quantification, J. Comput. Phys., 321, 1098-1109 (2016) · Zbl 1349.65021 |
| [81] | Zhang, C.; Shahbaba, B.; Zhao, H., Hamiltonian Monte Carlo acceleration using surrogate functions with random bases, Stat. Comput., 27, 6, 1473-1490 (2017) · Zbl 1384.65008 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
