Itô-SDE MCMC method for Bayesian characterization of errors associated with data limitations in stochastic expansion methods for uncertainty quantification

Abstract This paper is concerned with the characterization and the propagation of errors associated with data limitations in polynomial-chaos-based stochastic methods for uncertainty quantification. Such an issue can arise in uncertainty quantification when only a limited amount of data is available. When the available information does not suffice to accurately determine the probability distributions that must be assigned to the uncertain variables, the Bayesian method for assigning these probability distributions becomes attractive because it allows the stochastic model to account explicitly for insufficiency of the available information. In previous work, such applications of the Bayesian method had already been implemented by using the Metropolis–Hastings and Gibbs Markov Chain Monte Carlo (MCMC) methods. In this paper, we present an alternative implementation, which uses an alternative MCMC method built around an Ito stochastic differential equation (SDE) that is ergodic for the Bayesian posterior. We draw together from the mathematics literature a number of formal properties of this Ito SDE that lend support to its use in the implementation of the Bayesian method, and we describe its discretization, including the choice of the free parameters, by using the implicit Euler method. We demonstrate the proposed methodology on a problem of uncertainty quantification in a complex nonlinear engineering application relevant to metal forming.

[1]  M. Girolami,et al.  Riemann manifold Langevin and Hamiltonian Monte Carlo methods , 2011, Journal of the Royal Statistical Society: Series B (Statistical Methodology).

[2]  G. M.,et al.  Partial Differential Equations I , 2023, Applied Mathematical Sciences.

[3]  A. Kennedy,et al.  Hybrid Monte Carlo , 1988 .

[4]  R. Tweedie,et al.  Exponential convergence of Langevin distributions and their discrete approximations , 1996 .

[5]  G. A. Pavliotis,et al.  Exponential return to equilibrium for hypoelliptic quadratic systems , 2011, 1106.2326.

[6]  Jian Wang,et al.  Strong Feller Continuity of Feller Processes and Semigroups , 2010, 1006.5286.

[7]  Kai Liu.  Stochastic Stability of Differential Equations in Abstract Spaces , 2022 .

[8]  O. L. Maître,et al.  Spectral Methods for Uncertainty Quantification: With Applications to Computational Fluid Dynamics , 2010 .

[9]  G. Pavliotis.  Stochastic Processes and Applications: Diffusion Processes, the Fokker-Planck and Langevin Equations , 2014 .

[10]  Christian Soize,et al.  Random Fields and Uncertainty Quantification in Solid Mechanics of Continuum Media , 2017 .

[11]  Luc Rey-Bellet,et al.  Ergodic properties of Markov processes , 2006 .

[12]  Roger G. Ghanem,et al.  On the construction and analysis of stochastic models: Characterization and propagation of the errors associated with limited data , 2006, J. Comput. Phys..

[13]  Christian Soize,et al.  Time-domain formulation in computational dynamics for linear viscoelastic media with model uncertainties and stochastic excitation , 2012, Comput. Math. Appl..

[14]  P. A. P. Moran,et al.  Statistical inference with bivariate gamma distributions , 1969 .

[15]  J. Doob.  Stochastic processes , 1953 .

[16]  Roger G. Ghanem,et al.  Identification of Bayesian posteriors for coefficients of chaos expansions , 2010, J. Comput. Phys..

[17]  B. Leimkuhler,et al.  Molecular Dynamics: With Deterministic and Stochastic Numerical Methods , 2015 .

[18]  Carolyn Pillers Dobler.  Mathematical Statistics: Basic Ideas and Selected Topics , 2002 .

[19]  Christian Soize,et al.  Itô SDE-based Generator for a Class of Non-Gaussian Vector-valued Random Fields in Uncertainty Quantification , 2014, SIAM J. Sci. Comput..

[20]  Habib N. Najm,et al.  Stochastic spectral methods for efficient Bayesian solution of inverse problems , 2005, J. Comput. Phys..

[21]  Christian Soize,et al.  Construction of probability distributions in high dimension using the maximum entropy principle: Applications to stochastic processes, random fields and random matrices , 2008 .

[22]  U. Faber.  Asymptotics In Statistics Some Basic Concepts , 2016 .

[23]  Luc Papeleux,et al.  Finite element simulation of springback in sheet metal forming , 2002 .

[24]  Radford M. Neal.  MCMC Using Hamiltonian Dynamics , 2011, 1206.1901.

[25]  Jonathan C. Mattingly,et al.  Ergodicity for SDEs and approximations: locally Lipschitz vector fields and degenerate noise , 2002 .

[26]  Van Der Vaart,et al.  The Bernstein-Von-Mises theorem under misspecification , 2012 .

[27]  Jean-Philippe Ponthot,et al.  AN OVERVIEW OF NONINTRUSIVE CHARACTERIZATION, PROPAGATION, AND SENSITIVITY ANALYSIS OF UNCERTAINTIES IN COMPUTATIONAL MECHANICS , 2014 .

[28]  Brian Staber,et al.  Approximate Solutions of Lagrange Multipliers for Information-Theoretic Random Field Models , 2015, SIAM/ASA J. Uncertain. Quantification.

[29]  David B. Dunson,et al.  Bayesian Data Analysis , 2010 .

[30]  Christian Soize,et al.  Polynomial Chaos Expansion of a Multimodal Random Vector , 2015, SIAM/ASA J. Uncertain. Quantification.

[31]  Dudley,et al.  Real Analysis and Probability: Integration , 2002 .

[32]  Vladas Sidoravicius,et al.  Stochastic Processes and Applications , 2007 .

[33]  T. Lelièvre,et al.  Free Energy Computations: A Mathematical Perspective , 2010 .

[34]  Christian Soize,et al.  The Fokker-Planck Equation for Stochastic Dynamical Systems and Its Explicit Steady State Solutions , 1994, Series on Advances in Mathematics for Applied Sciences.

[35]  Jerzy Zabczyk,et al.  Ergodicity for Infinite Dimensional Systems: Appendices , 1996 .

[36]  L. Hörmander.  The analysis of linear partial differential operators , 1990 .