Počet záznamů: 1
Efficient Implementation of the Bayesian Inversion by MCMC with Acceleration of Posterior Sampling Using Surrogate Models
- 1.
SYSNO ASEP 0543700 Druh ASEP C - Konferenční příspěvek (mezinárodní konf.) Zařazení RIV D - Článek ve sborníku Název Efficient Implementation of the Bayesian Inversion by MCMC with Acceleration of Posterior Sampling Using Surrogate Models Tvůrce(i) Domesová, Simona (UGN-S) ORCID, SAI, RID
Béreš, Michal (UGN-S) ORCID, RID, SAI
Blaheta, Radim (UGN-S) RID, SAI, ORCIDCelkový počet autorů 3 Zdroj.dok. Lecture Notes in Civil Engineering, Challenges and Innovations in Geomechanics. - Cham : Springer, 2021 / Barla M. ; Di Donna A. ; Sterpi D. - ISSN 2366-2557 - ISBN 978-3-030-64513-7 Rozsah stran (2021), s. 846-853 Poč.str. 8 s. Forma vydání Online - E Akce International Conference of the International Association for Computer Methods and Advances in Geomechanics /16./ Datum konání 05.05.2021 - 08.05.2021 Místo konání Turin Země IT - Itálie Typ akce WRD Jazyk dok. eng - angličtina Země vyd. CH - Švýcarsko Klíč. slova Bayesian inversion ; posterior sampling ; delayed acceptance Metropolis-Hastings algorithm ; surrogate model ; inverse problems in hydro-mechanics Vědní obor RIV JN - Stavebnictví Obor OECD Applied mathematics CEP TK02010118 GA TA ČR - Technologická agentura ČR Způsob publikování Omezený přístup Institucionální podpora UGN-S - RVO:68145535 EID SCOPUS 85101546848 DOI 10.1007/978-3-030-64514-4_91 Anotace The contribution is motivated by the Bayesian approach to the solution of material identification problems which frequently appear in geo-engineering. We shall consider the cases with associated forward model describing flow in porous media with or without fractures as well as coupled hydro-mechanical processes. When assuming uncertainties in observed data, the use of the Bayesian inversion is natural. In comparison to deterministic methods, which lead only to a point estimate of the identified parameters, the Bayesian approach provides their probability distribution. The implementation of the Bayesian inversion is realized via Markov Chain Monte Carlo methods. The paper aims at the acceleration of the posterior sampling using a surrogate model that provides a polynomial approximation of the full forward model. The sampling procedure is based on the delayed acceptance Metropolis-Hastings (DAMH) algorithm. Therefore, for each proposed sample, the acceptance decision contains a preliminary step, which works only with an approximated posterior distribution constructed using the surrogate model. Furthermore, the approximated posterior distribution is being updated using new snapshots obtained during the sampling process. The posterior distribution updates are realized via updates of the surrogate model. The application of the described approach is shown through several model examples including flow in porous media with fractures and hydro-mechanical coupling. Pracoviště Ústav geoniky Kontakt Lucie Gurková, lucie.gurkova@ugn.cas.cz, Tel.: 596 979 354 Rok sběru 2022 Elektronická adresa https://link.springer.com/chapter/10.1007%2F978-3-030-64514-4_91
Počet záznamů: 1