Abstract
We are concerned with the efficiency of stochastic gradient estimation methods for large-scale nonlinear optimization in the presence of uncertainty. These methods aim to estimate an approximate gradient from a limited number of random input vector samples and corresponding objective function values. Ensemble methods usually employ Gaussian sampling to generate the input samples. It is known from the optimal design theory that the quality of sample-based approximations is affected by the distribution of the samples. We therefore evaluate six different sampling strategies to optimization of a high-dimensional analytical benchmark optimization problem, and, in a second example, to optimization of oil reservoir management strategies with and without geological uncertainty. The effectiveness of the sampling strategies is analyzed based on the quality of the estimated gradient, the final objective function value, the rate of the convergence, and the robustness of the gradient estimate. Based on the results, an improved version of the stochastic simplex approximate gradient method is proposed based on UE(s2) sampling designs for supersaturated cases that outperforms all alternative approaches. We additionally introduce two new strategies that outperform the UE(s2) designs previously suggested in the literature.
Article PDF
Similar content being viewed by others
References
de Aguiar, P.F., Bourguignon, B., Khots, M.S., Massart, D.L., Phan-Than-Luu, R.: D-optimal designs. Chemom. Intell. Lab. Syst. 30(2), 199–210 (1995)
Booth, K.H.V., Cox, D.R.: Some systematic supersaturated designs. Technometrics 4, 489–495 (1962)
Brouwer, D.R., Jansen, J.D.: Dynamic optimization of waterflooding with smart wells using optimal control theory. SPE J. 9(4), 391–402 (2004)
Caflisch, R.E.: Monte Carlo and quasi-Monte Carlo methods. Acta Numerica 7, 1–49 (1998). https://doi.org/10.1017/S0962492900002804
Cavazzuti, M: Optimization methods: from theory to design scientific and technological aspects in mechanics. Springer, ISBN: 978-3-642-31186-4 (2013)
Chen, Y.: Efficient ensemble based reservoir management/ PhD Thesis. University of Oklahoma, USA (2008)
Custodio, A., Vicente, L.: Using sampling and simplex derivatives in pattern search methods. SIAM J. Optim. 18(2), 537–555 (2007)
Diego, G., Loyola, R., Pedergnana, M., Garcia, S.G.: Smart sampling and incremental function learning for very large high dimensional data. Neural Netw. 78, 75–87 (2016)
Dixon, L.C.W., Mills, D.J.: Effect of rounding errors on the variable metric method. J. Optim. Theory Appl. 80(1), 175–179 (1994)
Do, S., Reynolds, A.: Theoretical connections between optimization algorithms based on an approximate gradient. Comput. Geosci. 17(6), 959–973 (2013). https://doi.org/10.1007/s10596-013-9368-9
Fonseca, R.M., Stordal, A.S., Leeuwenburgh, O., den Hof, P.M.J.V., Jansen, J.D.: Robust ensemble-based multi-objective optimization. In: Proceedings of the 14th European Conference on the Mathematics of Oil Recovery (ECMOR XIV), 8-11 September, Catania (2014)
Fonseca, R.M., Kahrobaei, S., Van Gastel, L.J.T., Leeuwenburgh, O., Jansen, J.D.: Quantification of the impact of ensemble size on the quality of an ensemble gradient using principles of hypothesis testing. Paper SPE 173236-MS presented at the SPE Reservoir Simulation Symposium, Houston, USA, 23-25 February (2015)
Fonseca, R.M., Leeuwenburgh, O., Van den Hof, P.M.J., Jansen, J.D.: Improving the ensemble optimization method through covariance matrix adaptation (cma-enopt). SPE J. 20(1), 155–168 (2015). https://doi.org/10.2118/163657-PA
Fonseca, R.M., Chen, B., Jansen, J.D., Reynolds, A.: A stochastic simplex approximate gradient (stosag) for optimization under uncertainty. Int. J. Numer. Methods Eng. 109(13), 1756–1776 (2016)
Gilmour, S.G.: Factor screening via supersaturated designs. In: Dean, A., Lewis, S. (eds.) Screening: methods for experimentation in industry, drug discovery and genetics, vol. 8, pp 169–190. Springer, New York (2006)
Van den Hof, P.M.J., Jansen, J.D., Van Essen, G., Bosgra, O.H.: Model based control and optimization of large scale physical systems-challenges in reservoir engineering. Chinese Control and Decision Conference, June 17-19 2009. Guilin, China (2009)
Jansen, J.D., Bosgra, O.H., Van den Hof, P.M.J.: Model-based control of multiphase flow in subsurface oil reservoirs. J. Process. Control. 18(9), 846–855 (2008)
Jansen, J.D., Fonseca, R.M., Kahrobaei, S.S., Siraj, M.M., Van Essen, G.M., Van den Hof, P.M.J.: The egg model (research note). http://data.4tu.nl/repository/uuid:916c86cd-3558-4672-829a-105c62985ab2, online accessed July 2017 (2013)
Jones, B., Majumdar, D.: Optimal supersaturated designs. J. Am. Stat. Assoc. 109(508), 1592–1600 (2014). https://doi.org/10.1080/01621459.2014.938810
Kelly, C.T.: Detection and remediation of stagnation in the nelder-mead algorithm using a sufficient decrease condition. SIAM J Optim 10(1), 43–55 (1999)
Lin, D.K.J.: A new class of supersaturated designs. Technometrics 35, 28–31 (1993)
McKay, M., Beckman, R., Conover, W.: A comparison of three methods for selecting values of input variables in the analysis of output from a computer code. Technometrics 42(1), 55–61 (1979)
Morokoff, W.J., Caflisch, R.E.: Quasi-Monte Carlo integration. Comput. Phys. 122(2), 218–230 (1995)
Niederreiter, H.: Quasi-Monte Carlo methods and pseudo-random numbers. Bull. Am. Math. Soc. 84(6), 957–1041 (1978)
Niederreiter, H.: Low-discrepancy and low-dispersion sequences. J. Number Theory 30, 51–70 (1988)
Okano, H., Koda, M.: An optimization algorithm based on stochastic sensitivity analysis for noisy objective landscapes. SPE J. 79(2), 245–252 (2003)
Owen, A.B.: Monte carlo variance of scrambled net quadrature. SIAM J. Numer. Anal. 34(5), 1884–1910 (1997)
Owen, A.B.: Monte Carlo theory, methods and examples. http://statweb.stanford.edu/owen/mc (2013)
Paskov, S.H., Traub, J.: Faster evaluation of financial derivatives. J. Portfolio Manag. 22, 113–120 (1995)
Raniolo, S., Dovera, L., Cominelli, A., Callegaro, C., Masserano, F.: History matching and polymer injection optimization in a mature field using the ensemble Kalman filter. In: Proc 17th European Symposium Improved Oil Recovery, St Petersburg, Russia, 16-18 April (2013)
Sarma, P., Chen, W.: Improved estimation of the stochastic gradient with quasi-Monte Carlo methods. In: Proc 14th European Conference on the Mathematics of Oil Recovery (ECMOR XIV), Catania, Italy, 8-11 September (2014)
Sarma, P., Chen, W., Aziz, K., Durlofsky, L.: Production optimization with adjoint models under non-linear control-state path inequality constraints. SPE Reserv. Eval. Eng., 326–339 (2008)
Spall, J.: Multivariate stochastic approximation using a simultaneous perturbation gradient approximation. IEEE Trans. Autom. Control 37(3), 332–341 (1992)
Stordal, A.S., Szklarz, S.P., Leeuwenburgh, O.: A theoretical look at ensemble-based optimization in reservoir management. Math. Geosci. 48(4), 399–417 (2016). https://doi.org/10.1007/s11004-015-9598-6
Van Essen, G., Zandvliet, M., Van den Hof, P., Bosgra, O., Jansen, J.D.: Robust waterflooding optimization of multiple geological scenarios. SPE J. 14(1), 202–210 (2009)
Wu, C.F.J.: Construction of supersaturated designs through partially aliased interactions. Biometrika 80, 661–669 (1993)
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
Ramaswamy, K.R., Fonseca, R.M., Leeuwenburgh, O. et al. Improved sampling strategies for ensemble-based optimization. Comput Geosci 24, 1057–1069 (2020). https://doi.org/10.1007/s10596-019-09914-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10596-019-09914-8