Abstract
The embedded ensemble propagation approach introduced in Phipps et al. (SIAM J. Sci. Comput. 39(2):C162, 2017) has been demonstrated to be a powerful means of reducing the computational cost of sampling-based uncertainty quantification methods, particularly on emerging computational architectures. A substantial challenge with this method however is ensemble-divergence, whereby different samples within an ensemble choose different code paths. This can reduce the effectiveness of the method and increase computational cost. Therefore grouping samples together to minimize this divergence is paramount in making the method effective for challenging computational simulations. In this work, a new grouping approach based on a surrogate for computational cost built up during the uncertainty propagation is developed and applied to model advection-diffusion problems where computational cost is driven by the number of (preconditioned) linear solver iterations. The approach is developed within the context of locally adaptive stochastic collocation methods, where a surrogate for the number of linear solver iterations, generated from previous levels of the adaptive grid generation, is used to predict iterations for subsequent samples, and group them based on similar numbers of iterations. The effectiveness of the method is demonstrated by applying it to highly anisotropic advection-dominated diffusion problems with a wide variation in solver iterations from sample to sample. It extends the parameter-based grouping approach developed in D’Elia et al. (SIAM/ASA J. Uncertain. Quantif. 6:87, 2017) to more general problems without requiring detailed knowledge of how the uncertain parameters affect the simulation’s cost, and is also less intrusive to the simulation code.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
Here, Ω is a set of realizations, \(\mathscr {F}\) is a σ-algebra of events and \(\mathbb {P}:\mathscr {F}\rightarrow [0,1]\) is a probability measure.
- 2.
- 3.
Note that here we allow the forcing term f to be sample dependent.
- 4.
Note that at each level, the number of samples is not usually evenly divisible by the ensemble size. To use a uniform ensemble size for all ensembles, samples are added by replicating the last sample in the last ensemble. This results in larger R values when the number of samples is small and the ensemble size is large, as can be seen in the results for R 1.
- 5.
Note that we do not include timing results in Table 4 since the ensemble implementation is currently not optimized for GMRES. A significant cost within GMRES is the orthogonalization of each new Krylov vector against the previous set of vectors, which in Trilinos is implemented through GEMV dense matrix-vector product BLAS routine. An optimized implementation of this routine for ensembles is currently being developed.
References
Babuška, I., Tempone, R., Zouraris, G.E.: Galerkin finite element approximations of stochastic elliptic partial differential equations. SIAM J. Numer. Anal. 42(2), 800–825 (2004)
Babuška, I., Nobile, F., Tempone, R.: A stochastic collocation method for elliptic partial differential equations with random input data. SIAM J. Numer. Anal. 45(3), 1005–1034 (2007)
Bäck, J., Nobile, F., Tamellini, L., Tempone, R.: Stochastic spectral Galerkin and collocation methods for PDEs with random coefficients: a numerical comparison. In: Spectral and High Order Methods for Partial Differential Equations, pp. 43–62. Springer, Berlin (2011)
Baker, C.G., Heroux, M.A.: Tpetra, and the use of generic programming in scientific computing. Sci. Program. 20(2), 115–128 (2012)
Bakr, M.H., Bandler, J.W., Madsen, K., Rayas-Sanchez, J.E., Søndergaard, J.: Space-mapping optimization of microwave circuits exploiting surrogate models. IEEE Trans. Microwave Theory Tech. 48(12), 2297–2306 (2000)
Bandler, J.W., Cheng, Q., Gebre-Mariam, D.H., Madsen, K., Pedersen, F., Sondergaard, J.: Em-based surrogate modeling and design exploiting implicit, frequency and output space mappings. In: Microwave Symposium Digest, 2003 IEEE MTT-S International, vol. 2, pp. 1003–1006. IEEE, Piscataway (2003)
Barth, A., Lang, A.: Multilevel Monte Carlo method with applications to stochastic partial differential equations. Int. J. Comput. Math. 89(18), 2479–2498 (2012)
Barth, A., Schwab, C., Zollinger, N.: Multi-level Monte Carlo finite element method for elliptic PDEs with stochastic coefficients. Numer. Math. 119(1), 123–161 (2011)
Barth, A., Lang, A., Schwab, C.: Multilevel Monte Carlo method for parabolic stochastic partial differential equations. BIT Numer. Math. 53(1), 3–27 (2013)
Bavier, E., Hoemmen, M., Rajamanickam, S., Thornquist, H.: Amesos2 and Belos: direct and iterative solvers for large sparse linear systems. Sci. Program. 20(3), 241–255 (2012)
Bichon, B.J., McFarland, J.M., Mahadevan, S.: Efficient surrogate models for reliability analysis of systems with multiple failure modes. Reliab. Eng. Syst. Saf. 96(10), 1386–1395 (2011)
Breitkopf, P., Coelho, R.F.: Multidisciplinary Design Optimization in Computational Mechanics. Wiley, Hoboken (2013)
Bungartz, H.J., Griebel, M.: Sparse grids. Acta Numer. 13(1), 147–269 (2004)
Cliffe, K.A., Giles, M.B., Scheichl, R., Teckentrup, A.L.: Multilevel Monte Carlo methods and applications to elliptic PDEs with random coefficients. Comput. Vis. Sci. 14(1), 3–15 (2011)
Cohen, A., DeVore, R., Schwab, C.: Analytic regularity and polynomial approximation of parametric and stochastic elliptic PDE’s. Anal. Appl. 9(01), 11–47 (2011)
Couckuyt, I., Declercq, F., Dhaene, T., Rogier, H., Knockaert, L.: Surrogate-based infill optimization applied to electromagnetic problems. Int. J. RF Microwave Comput. Aided Eng. 20(5), 492–501 (2010)
D’Elia, M., Edwards, H.C., Hu, J., Phipps, E., Rajamanickam, S.: Ensemble grouping strategies for embedded stochastic collocation methods applied to anisotropic diffusion problems. SIAM/ASA J. Uncertain. Quantif. 6, 87 (2017)
Donea, J., Huera, A.: Finite Element Methods for Flow Problems. Wiley, New York (2003)
Doostan, A., Owhadi. H.: A non-adapted sparse approximation of PDEs with stochastic inputs. J. Comput. Phys. 230(8), 3015–3034 (2011)
Ebeida, M.S., Mitchell, S.A., Swiler, L.P., Romero, V.J., Rushdi, A.A.: POF-darts: geometric adaptive sampling for probability of failure. Reliab. Eng. Syst. Saf. 155, 64–77 (2016)
Edwards, H.C., Sunderland, D., Porter, V., Amsler, C., Mish, S.: Manycore performance-portability: Kokkos multidimensional array library. Sci. Program. 20(2), 89–114 (2012)
Edwards, H.C., Trott, C.R., Sunderland, D.: Kokkos: enabling manycore performance portability through polymorphic memory access patterns. J. Parallel Distrib. Comput. 74, 3202–3216 (2014)
Fishman, G.S.: Monte Carlo: Concepts, Algorithms, and Applications. Springer Series in Operations Research. Springer, New York (1996)
Forrester, A.I.J., Bressloff, N.W., Keane, A.J.: Optimization using surrogate models and partially converged computational fluid dynamics simulations. Proc. R. Soc. Lond. A Math. Phys. Eng. Sci. 462, 2177–2204 (2006). The Royal Society
Frauenfelder, P., Schwab, C., Todor, R.A.: Finite elements for elliptic problems with stochastic coefficients. Comput. Methods Appl. Mech. Eng. 194(2), 205–228 (2005)
Galindo, D., Jantsch, P., Webster, C.G., Zhang, G.: Accelerating stochastic collocation methods for PDEs with random input data. Technical Report TM–2015/219, Oak Ridge National Laboratory (2015)
Ganapathysubramanian, B., Zabaras, N.: Sparse grid collocation schemes for stochastic natural convection problems. J. Comput. Phys. 225(1), 652–685 (2007)
Ghanem, R.G., Spanos, P.D.: Polynomial chaos in stochastic finite elements. J. Appl. Mech. 57, 197 (1990)
Ghanem, R.G., Spanos, P.D.: Stochastic Finite Elements: A Spectral Approach. Springer, New York (1991)
Giles, M.B.: Multilevel Monte Carlo path simulation. Oper. Res. 56(3), 607–617 (2008)
Griebel, M.: Adaptive sparse grid multilevel methods for elliptic PDEs based on finite differences. Computing 61(2), 151–179 (1998)
Gunzburger, M., Webster, C.G., Zhang, G.: An adaptive wavelet stochastic collocation method for irregular solutions of partial differential equations with random input data. In: Sparse Grids and Applications – Munich 2012, pp. 137–170. Springer International Publishing, Cham (2014)
Gunzburger, M.D., Webster, C.G., Zhang, G.: Stochastic finite element methods for partial differential equations with random input data. Acta Numer. 23, 521–650 (2014)
Hao, P., Wang, B., Li, G.: Surrogate-based optimum design for stiffened shells with adaptive sampling. AIAA J. 50(11), 2389–2407 (2012)
Helton, J.C., Davis, F.J.: Latin hypercube sampling and the propagation of uncertainty in analyses of complex systems. Reliab. Eng. Syst. Saf. 81, 23–69 (2003)
Heroux, M.A., Willenbring, J.M.: A new overview of the Trilinos project. Sci. Program. 20(2), 83–88 (2012)
Heroux, M.A., Bartlett, R.A., Howle, V.E., Hoekstra, R.J., Hu, J.J., Kolda, T.G., Lehoucq, R.B., Long, K.R., Pawlowski, R.P., Phipps, E.T., Salinger, A.G., Thornquist, H.K., Tuminaro, R.S., Willenbring, J.M., Williams, A.B., Stanley, K.S.: An overview of the Trilinos package. ACM Trans. Math. Softw. 31(3), 397 (2005)
Li, J., Xiu, D.: Evaluation of failure probability via surrogate models. J. Comput. Phys. 229(23), 8966–8980 (2010)
Li, J., Li, J., Xiu, D.: An efficient surrogate-based method for computing rare failure probability. J. Comput. Phys. 230(24), 8683–8697 (2011)
Loève, M.: Probability Theory I. Graduate Texts in Mathematics, 4th edn., vol. 45. Springer, New York (1977)
Loève, M.: Probability Theory II. Graduate Texts in Mathematics, 4th edn., vol. 46. Springer, New York (1978)
Mathelin, L., Gallivan, K.A.: A compressed sensing approach for partial differential equations with random input data. Commun. Comput. Phys. 12(04), 919–954 (2012)
McKay, M.D., Beckman, R.J., Conover, W.J.: A 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)
Metropolis, N., Ulam, S.: The Monte Carlo method. J. Am. Stat. Assoc. 44(247), 335–341 (1949)
Niederreiter, H.: Quasi-Monte Carlo methods and pseudo-random numbers. Bull. Amer. Math. Soc 84(6), 957–1041 (1978)
Nobile, F., Tempone, R., Webster, C.G.: A sparse grid stochastic collocation method for partial differential equations with random input data. SIAM J. Numer. Anal. 46(5), 2309–2345 (2008)
Nobile, F., Tempone, R., Webster, C.G.: An anisotropic sparse grid stochastic collocation method for partial differential equations with random input data. SIAM J. Numer. Anal. 46(5), 2411–2442 (2008)
Pawlowski, R.P., Phipps, E.T., Salinger, A.G.: Automating embedded analysis capabilities and managing software complexity in multiphysics simulation, Part I: Template-based generic programming. Sci. Program. 20, 197–219 (2012)
Pawlowski, R.P., Phipps, E.T., Salinger, A.G., Owen, S.J., Siefert, C.M., Staten, M.L.: Automating embedded analysis capabilities and managing software complexity in multiphysics simulation part II: application to partial differential equations. Sci. Program. 20, 327–345 (2012)
Phipps, E.T.: Stokhos Stochastic Galerkin Uncertainty Quantification Methods (2015). Available online: http://trilinos.org/packages/stokhos/
Phipps, E., D’Elia, M., Edwards, H.C., Hoemmen, M., Hu, J., Rajamanickam, S.: Embedded ensemble propagation for improving performance, portability and scalability of uncertainty quantification on emerging computational architectures. SIAM J. Sci. Comput. 39(2), C162 (2017)
Prokopenko, A., Hu, J.J., Wiesner, T.A., Siefert, C.M., Tuminaro, R.S.: MueLu user’s guide 1.0. Technical Report SAND2014-18874, Sandia National Laboratories (2014)
Razavi, S., Tolson, B.A., Burn, D.H.: Review of surrogate modeling in water resources. Water Resour. Res. 48(7), 7401 (2012)
Rikards, R., Abramovich, H., Auzins, J., Korjakins, A., Ozolinsh, O., Kalnins, K., Green, T.: Surrogate models for optimum design of stiffened composite shells. Compos. Struct. 63(2), 243–251 (2004)
Roman, L.J., Sarkis, M.: Stochastic Galerkin method for elliptic SPDEs: a white noise approach. Discrete Contin. Dynam. Syst. B 6(4), 941 (2006)
Smolyak, S.A.: Quadrature and interpolation formulas for tensor products of certain classes of functions. Dokl. Akad. Nauk SSSR 4, 240–243 (1963)
Stoyanov, M.: Hierarchy-direction selective approach for locally adaptive sparse grids. Technical Report TM–2013/384, Oak Ridge National Laboratory (2013)
Stoyanov, M., Webster, C.G.: A dynamically adaptive sparse grid method for quasi-optimal interpolation of multidimensional analytic functions. Technical Report TM–2015/341, Oak Ridge National Laboratory (2015)
Xiu, D.B., Hesthaven, J.S.: High-order collocation methods for differential equations with random inputs. SIAM J. Sci. Comput. 27(3), 1118–1139 (2005)
Xiu, D.B., Karniadakis, G.E.: The Wiener-Askey polynomial chaos for stochastic differential equations. SIAM J. Sci. Comput. 24(2), 619–644 (2002)
Acknowledgements
Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525. This paper describes objective technical results and analysis. Any subjective views or opinions that might be expressed in the paper do not necessarily represent the views of the U.S. Department of Energy or the United States Government.
This material is based upon work supported by the U.S. Department of Energy, Office of Science, and Office of Advanced Scientific Computing Research (ASCR), as well as the National Nuclear Security Administration, Advanced Technology Development and Mitigation program. This research used resources of the Oak Ridge Leadership Computing Facility, which is a DOE Office of Science User Facility.
The authors would like to thank Dr. Miro Stoyanov for useful conversations and for providing great support with TASMANIAN.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 National Technology & Engineering Solutions of Sandia, and E. Phipps, A. Rushdi, and M. S. Ebeida, under exclusive license to SIP AG, part of Springer Nature
About this chapter
Cite this chapter
D’Elia, M., Phipps, E., Rushdi, A., Ebeida, M.S. (2020). Surrogate-Based Ensemble Grouping Strategies for Embedded Sampling-Based Uncertainty Quantification. In: D'Elia, M., Gunzburger, M., Rozza, G. (eds) Quantification of Uncertainty: Improving Efficiency and Technology. Lecture Notes in Computational Science and Engineering, vol 137 . Springer, Cham. https://doi.org/10.1007/978-3-030-48721-8_3
Download citation
DOI: https://doi.org/10.1007/978-3-030-48721-8_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-48720-1
Online ISBN: 978-3-030-48721-8
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)