Non-Linear Approximation of Bayesian Update

Handle URI:
http://hdl.handle.net/10754/623696
Title:
Non-Linear Approximation of Bayesian Update
Authors:
Litvinenko, Alexander ( 0000-0001-5427-3598 )
Abstract:
We develop a non-linear approximation of expensive Bayesian formula. This non-linear approximation is applied directly to Polynomial Chaos Coefficients. In this way, we avoid Monte Carlo sampling and sampling error. We can show that the famous Kalman Update formula is a particular case of this update.
KAUST Department:
SRI Center for Uncertainty Quantification in Computational Science and Engineering, King Abdullah University of Science and Technology, Thuwal, Saudi Arabia
Conference/Event name:
ECRC Meeting
Issue Date:
23-Jun-2016
Type:
Presentation
Sponsors:
SRI UQ, ECRC KAUST
Appears in Collections:
Presentations

Full metadata record

DC FieldValue Language
dc.contributor.authorLitvinenko, Alexanderen
dc.date.accessioned2017-05-23T06:17:42Z-
dc.date.available2017-05-23T06:17:42Z-
dc.date.issued2016-06-23-
dc.identifier.urihttp://hdl.handle.net/10754/623696-
dc.description.abstractWe develop a non-linear approximation of expensive Bayesian formula. This non-linear approximation is applied directly to Polynomial Chaos Coefficients. In this way, we avoid Monte Carlo sampling and sampling error. We can show that the famous Kalman Update formula is a particular case of this update.en
dc.description.sponsorshipSRI UQ, ECRC KAUSTen
dc.subjectLorenz 84en
dc.subjectBayesian Updateen
dc.subjectPCEen
dc.subjectBayesian surrogateen
dc.subjectupdate of PCE coefficientsen
dc.subjectposterioren
dc.titleNon-Linear Approximation of Bayesian Updateen
dc.typePresentationen
dc.contributor.departmentSRI Center for Uncertainty Quantification in Computational Science and Engineering, King Abdullah University of Science and Technology, Thuwal, Saudi Arabiaen
dc.conference.dateJune 2016en
dc.conference.nameECRC Meetingen
dc.conference.locationKAUSTen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.