Estimation of distributions via multilevel Monte Carlo with stratified sampling

Taverniers, Søren; Tartakovsky, Daniel M.

doi:10.1016/j.jcp.2020.109572

Estimation of distributions via multilevel Monte Carlo with stratified sampling

Journal Article · Thu Jul 09 00:00:00 EDT 2020 · Journal of Computational Physics

DOI:https://doi.org/10.1016/j.jcp.2020.109572· OSTI ID:1852855

Taverniers, Søren ^[1]; ^[2]

Stanford Univ., CA (United States). Dept. of Energy Resources Engineering; Stanford Univ., CA (United States)
Stanford Univ., CA (United States). Dept. of Energy Resources Engineering

We design and implement a novel algorithm for computing a multilevel Monte Carlo (MLMC) estimator of the joint cumulative distribution function (CDF) of a vector-valued quantity of interest in problems with random input parameters and initial conditions. Our approach combines MLMC with stratified sampling of the input sample space by replacing standard Monte Carlo at each level with stratified Monte Carlo initialized with proportionally allocated samples. We show that the resulting stratified MLMC (sMLMC) algorithm is more efficient than its standard MLMC counterpart due to the additional variance reduction provided by the stratification of the random parameter's domain, especially at the coarsest levels. Additional computational cost savings are obtained by smoothing the indicator function with a Gaussian kernel, which proves to be an efficient and robust alternative to recently developed polynomial-based techniques.

View Accepted Manuscript (DOE)

Research Organization:: Stanford Univ., CA (United States)

Sponsoring Organization:: USDOE Office of Science (SC); Defense Advanced Research Project Agency (DARPA); US Air Force Office of Scientific Research (AFOSR)

Grant/Contract Number:: SC0019130

OSTI ID:: 1852855

Alternate ID(s):: OSTI ID: 1638394

Journal Information:: Journal of Computational Physics, Journal Name: Journal of Computational Physics Journal Issue: C Vol. 419; ISSN 0021-9991

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (36)

The method of distributions for dispersive transport in porous media with uncertain hydraulic properties: CDF METHOD FOR DISPERSIVE TRANSPORT Boso, Francesca; Tartakovsky, Daniel M. Water Resources Research, Vol. 52, Issue 6 https://doi.org/10.1002/2016wr018745	journal	June 2016
Coupling Importance Sampling and Multilevel Monte Carlo using Sample Average Approximation Kebaier, Ahmed; Lelong, Jérôme Methodology and Computing in Applied Probability, Vol. 20, Issue 2 https://doi.org/10.1007/s11009-017-9579-y	journal	July 2017
Impact of parametric uncertainty on estimation of the energy deposition into an irradiated brain tumor Taverniers, Søren; Tartakovsky, Daniel M. Journal of Computational Physics, Vol. 348 https://doi.org/10.1016/j.jcp.2017.07.008	journal	November 2017
Geostatistical modelling of uncertainty in soil science Goovaerts, P. Geoderma, Vol. 103, Issue 1-2 https://doi.org/10.1016/s0016-7061(01)00067-2	journal	September 2001
Optimal Model Management for Multifidelity Monte Carlo Estimation Peherstorfer, Benjamin; Willcox, Karen; Gunzburger, Max SIAM Journal on Scientific Computing, Vol. 38, Issue 5 https://doi.org/10.1137/15m1046472	journal	January 2016
Multilevel Quasi-Monte Carlo Methods for Lognormal Diffusion Problems Kuo, Frances Y.; Scheichl, Robert; Schwab, Christoph arXiv https://doi.org/10.48550/arxiv.1507.01090	preprint	January 2015
The method of distributions for dispersive transport in porous media with uncertain hydraulic properties: CDF METHOD FOR DISPERSIVE TRANSPORT Boso, Francesca; Tartakovsky, Daniel M. Water Resources Research, Vol. 52, Issue 6 https://doi.org/10.1002/2016WR018745	journal	June 2016
An improved multilevel Monte Carlo method for estimating probability distribution functions in stochastic oil reservoir simulations: AN IMPROVED MLMC METHOD Lu, Dan; Zhang, Guannan; Webster, Clayton Water Resources Research, Vol. 52, Issue 12 https://doi.org/10.1002/2016WR019475	journal	December 2016
Monte Carlo Complexity of Global Solution of Integral Equations Heinrich, S. Journal of Complexity, Vol. 14, Issue 2 https://doi.org/10.1006/jcom.1998.0471	journal	June 1998
Advances in prospect theory: Cumulative representation of uncertainty Tversky, Amos; Kahneman, Daniel Journal of Risk and Uncertainty, Vol. 5, Issue 4 https://doi.org/10.1007/BF00122574	journal	October 1992
Stochastic analysis of effective rate constant for heterogeneous reactions Lichtner, P. C.; Tartakovsky, D. M. Stochastic Environmental Research and Risk Assessment (SERRA), Vol. 17, Issue 6 https://doi.org/10.1007/s00477-003-0163-3	journal	December 2003
Multilevel Monte Carlo methods and applications to elliptic PDEs with random coefficients Cliffe, K. A.; Giles, M. B.; Scheichl, R. Computing and Visualization in Science, Vol. 14, Issue 1 https://doi.org/10.1007/s00791-011-0160-x	journal	January 2011
Geostatistical modelling of uncertainty in soil science Goovaerts, P. Geoderma, Vol. 103, Issue 1-2 https://doi.org/10.1016/S0016-7061(01)00067-2	journal	September 2001
Estimation of small failure probabilities in high dimensions by subset simulation Au, Siu-Kui; Beck, James L. Probabilistic Engineering Mechanics, Vol. 16, Issue 4 https://doi.org/10.1016/S0266-8920(01)00019-4	journal	October 2001
Assessment and management of risk in subsurface hydrology: A review and perspective Tartakovsky, Daniel M. Advances in Water Resources, Vol. 51 https://doi.org/10.1016/j.advwatres.2012.04.007	journal	January 2013
An efficient Bayesian data-worth analysis using a multilevel Monte Carlo method Lu, Dan; Ricciuto, Daniel; Evans, Katherine Advances in Water Resources, Vol. 113 https://doi.org/10.1016/j.advwatres.2018.01.024	journal	March 2018
Multilevel Monte Carlo for two phase flow and Buckley–Leverett transport in random heterogeneous porous media Müller, Florian; Jenny, Patrick; Meyer, Daniel W. Journal of Computational Physics, Vol. 250 https://doi.org/10.1016/j.jcp.2013.03.023	journal	October 2013
Solver-based vs. grid-based multilevel Monte Carlo for two phase flow and transport in random heterogeneous porous media Müller, Florian; Meyer, Daniel W.; Jenny, Patrick Journal of Computational Physics, Vol. 268 https://doi.org/10.1016/j.jcp.2014.02.047	journal	July 2014
Approximation of probability density functions by the Multilevel Monte Carlo Maximum Entropy method Bierig, Claudio; Chernov, Alexey Journal of Computational Physics, Vol. 314 https://doi.org/10.1016/j.jcp.2016.03.027	journal	June 2016
Moment Differential Equations for Flow in Highly Heterogeneous Porous Media Winter, C. L.; Tartakovsky, D. M.; Guadagnini, A. Surveys in Geophysics, Vol. 24, Issue 1, p. 81-106 https://doi.org/10.1023/A:1022277418570	journal	January 2003
Probability density functions for advective‐reactive transport with uncertain reaction rates Tartakovsky, Daniel M.; Dentz, Marco; Lichtner, Peter C. Water Resources Research, Vol. 45, Issue 7 https://doi.org/10.1029/2008WR007383	journal	July 2009
Efficient Monte Carlo With Graph‐Based Subsurface Flow and Transport Models O'Malley, D.; Karra, S.; Hyman, J. D. Water Resources Research, Vol. 54, Issue 5 https://doi.org/10.1029/2017WR022073	journal	May 2018
Multi-level methods and approximating distribution functions Wilson, D.; Baker, R. E. AIP Advances, Vol. 6, Issue 7 https://doi.org/10.1063/1.4960118	journal	July 2016
Multi-scale shock-to-detonation simulation of pressed energetic material: A meso-informed ignition and growth model Sen, O.; Rai, N. K.; Diggs, A. S. Journal of Applied Physics, Vol. 124, Issue 8 https://doi.org/10.1063/1.5046185	journal	August 2018
Sparse tensor multi-level Monte Carlo finite volume methods for hyperbolic conservation laws with random initial data Mishra, S.; Schwab, Ch. Mathematics of Computation, Vol. 81, Issue 280 https://doi.org/10.1090/S0025-5718-2012-02574-9	journal	April 2012
Multilevel Quasi-Monte Carlo methods for lognormal diffusion problems Kuo, Frances Y.; Scheichl, Robert; Schwab, Christoph Mathematics of Computation, Vol. 86, Issue 308 https://doi.org/10.1090/mcom/3207	journal	March 2017
Stochastic Collocation Methods for Nonlinear Parabolic Equations with Random Coefficients Barajas-Solano, David A.; Tartakovsky, Daniel M. SIAM/ASA Journal on Uncertainty Quantification, Vol. 4, Issue 1 https://doi.org/10.1137/130930108	journal	January 2016
Multilevel Monte Carlo Approximation of Distribution Functions and Densities Giles, Michael B.; Nagapetyan, Tigran; Ritter, Klaus SIAM/ASA Journal on Uncertainty Quantification, Vol. 3, Issue 1 https://doi.org/10.1137/140960086	journal	January 2015
A Multilevel Monte Carlo Method for Computing Failure Probabilities Elfverson, Daniel; Hellman, Fredrik; Målqvist, Axel SIAM/ASA Journal on Uncertainty Quantification, Vol. 4, Issue 1 https://doi.org/10.1137/140984294	journal	January 2016
Multilevel Estimation of Rare Events Ullmann, Elisabeth; Papaioannou, Iason SIAM/ASA Journal on Uncertainty Quantification, Vol. 3, Issue 1 https://doi.org/10.1137/140992953	journal	January 2015
Optimal Model Management for Multifidelity Monte Carlo Estimation Peherstorfer, Benjamin; Willcox, Karen; Gunzburger, Max SIAM Journal on Scientific Computing, Vol. 38, Issue 5 https://doi.org/10.1137/15M1046472	journal	January 2016
Multilevel Monte Carlo Approximation of Functions Krumscheid, S.; Nobile, F. SIAM/ASA Journal on Uncertainty Quantification, Vol. 6, Issue 3 https://doi.org/10.1137/17M1135566	journal	January 2018
Eulerian Moment Equations for 2-D Stochastic Immiscible Flow Jarman, Kenneth D.; Russell, Thomas F. Multiscale Modeling & Simulation, Vol. 1, Issue 4 https://doi.org/10.1137/S1540345902413176	journal	January 2003
Effective Ion Diffusion in Charged Nanoporous Materials Zhang, Xuan; Tartakovsky, Daniel M. Journal of The Electrochemical Society, Vol. 164, Issue 4 https://doi.org/10.1149/2.0491704jes	journal	January 2017
Remarks on Some Nonparametric Estimates of a Density Function Rosenblatt, Murray The Annals of Mathematical Statistics, Vol. 27, Issue 3 https://doi.org/10.1214/aoms/1177728190	journal	September 1956
Multilevel Monte Carlo Path Simulation Giles, Michael B. Operations Research, Vol. 56, Issue 3 https://doi.org/10.1287/opre.1070.0496	journal	June 2008

Cited By (2)

GINNs: Graph-Informed Neural Networks for multiscale physics Hall, Eric J.; Taverniers, Søren; Katsoulakis, Markos A. Journal of Computational Physics, Vol. 433 https://doi.org/10.1016/j.jcp.2021.110192	journal	May 2021
UNSW-NB15 Computer Security Dataset: Analysis through Visualization Zoghi, Zeinab; Serpen, Gursel arXiv https://doi.org/10.48550/arxiv.2101.05067	preprint	January 2021

Similar Records

Forward and inverse uncertainty quantification using multilevel Monte Carlo algorithms for an elliptic non-local equation

Journal Article · Thu Dec 31 23:00:00 EST 2015 · International Journal for Uncertainty Quantification · OSTI ID:1342665

Advanced Multilevel Monte Carlo Methods

Journal Article · Mon Mar 02 23:00:00 EST 2020 · International Statistical Review · OSTI ID:1862171

An improved multilevel Monte Carlo method for estimating probability distribution functions in stochastic oil reservoir simulations

Journal Article · Thu Dec 29 23:00:00 EST 2016 · Water Resources Research · OSTI ID:1338482

Related Subjects

71 CLASSICAL AND QUANTUM MECHANICS
GENERAL PHYSICS
97 MATHEMATICS AND COMPUTING
Computer Science
Cumulative distribution function
Kernel density estimation
Multilevel Monte Carlo
Physics
Stratified sampling

Estimation of distributions via multilevel Monte Carlo with stratified sampling

Citation Formats

References (36)

Cited By (2)

Similar Records

Related Subjects