CaltechAUTHORS
  A Caltech Library Service

Bayesian inference with Subset Simulation: Strategies and improvements

Betz, Wolfgang and Papaioannou, Iason and Beck, James L. and Straub, Daniel (2018) Bayesian inference with Subset Simulation: Strategies and improvements. Computer Methods in Applied Mechanics and Engineering, 331 . pp. 72-93. ISSN 0045-7825. http://resolver.caltech.edu/CaltechAUTHORS:20171206-101213438

[img] PDF - Accepted Version
See Usage Policy.

1695Kb

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20171206-101213438

Abstract

Bayesian Updating with Structural reliability methods (BUS) reinterprets the Bayesian updating problem as a structural reliability problem; i.e. a rare event estimation. The BUS approach can be considered an extension of rejection sampling, where a standard uniform random variable is added to the space of random variables. Each generated sample from this extended random variable space is accepted if the realization of the uniform random variable is smaller than the likelihood function scaled by a constant c. The constant c has to be selected such that 1∕c is not smaller than the maximum of the likelihood function, which, however, is typically unknown a-priori. A c chosen too small will have negative impact on the efficiency of the BUS approach when combined with sampling-based reliability methods. For the combination of BUS with Subset Simulation, we propose an approach, termed aBUS, for adaptive BUS, that does not require c as input. The proposed algorithm requires only minimal modifications of standard BUS with Subset Simulation. We discuss why aBUS produces samples that follow the posterior distribution –even if 1∕c is selected smaller than the maximum of the likelihood function. The performance of aBUS in terms of the computed evidence required for Bayesian model class selection and in terms of the produced posterior samples is assessed numerically for different example problems. The combination of BUS with Subset Simulation (and aBUS in particular) is well suited for problems with many uncertain parameters and for Bayesian updating of models where it is computationally demanding to evaluate the likelihood function.


Item Type:Article
Related URLs:
URLURL TypeDescription
https://doi.org/10.1016/j.cma.2017.11.021DOIArticle
http://www.sciencedirect.com/science/article/pii/S0045782517307211PublisherArticle
Additional Information:© 2017 Elsevier B.V. Received 10 May 2017, Revised 8 November 2017, Accepted 13 November 2017, Available online 2 December 2017. We acknowledge the support of the Technische Universität München–Institute for Advanced Study, funded by the German Excellence Initiative.
Funders:
Funding AgencyGrant Number
Technische Universität MünchenUNSPECIFIED
Deutsche Forschungsgemeinschaft (DFG)UNSPECIFIED
Subject Keywords:Bayesian updating; Bayesian model class selection; Subset Simulation; Structural reliability; MCMC; BUS
Record Number:CaltechAUTHORS:20171206-101213438
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20171206-101213438
Official Citation:Wolfgang Betz, Iason Papaioannou, James L. Beck, Daniel Straub, Bayesian inference with Subset Simulation: Strategies and improvements, Computer Methods in Applied Mechanics and Engineering, Volume 331, 1 April 2018, Pages 72-93, ISSN 0045-7825, https://doi.org/10.1016/j.cma.2017.11.021. (https://www.sciencedirect.com/science/article/pii/S0045782517307211)
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:83722
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:08 Dec 2017 03:35
Last Modified:16 Dec 2017 00:25

Repository Staff Only: item control page