BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//EURODYN//Schedule//DE
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Berlin
X-LIC-LOCATION:Europe/Berlin
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700329T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701025T030000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:152@eurodyn2026.org
DTSTAMP:20261026T144037Z
DTSTART;TZID=Europe/Berlin:20260928T165000
DTEND;TZID=Europe/Berlin:20260928T171000
SUMMARY:Active Bayesian Inference via Gaussian Process Surrogates for Par
 ameter and Error Estimation
LOCATION:Room F107
DESCRIPTION:MS 28\n\nPresenter: Acevedo Farías\, Cristóbal Hernán\n\nAuth
 ors: Acevedo\, Cristóbal H.\; Valdebenito\, Marcos A.\; Keßler\, Sylvia\
 ; Faes\, Matthias G.R.\n\nChair: Faes\, Matthias\n\nAbstract: Uncertaint
 ies in the parameters of an engineering model often lead to differences 
 between its predictions and experimental data that was observed on the p
 hysical structure. To deal with these uncertainties\, updating the param
 eters given the data is a possible approach. Bayesian inference provides
  a framework for this purpose\, which also incorporates the prior inform
 ation about the parameters. One drawback of this method is the high comp
 utational cost of the evaluations of the model responses\, which are nec
 essary for the inference process. This work proposes a solution to this 
 high computational cost that includes Gaussian Process (GP) regression t
 o approximate the model responses. This GP surrogate is trained with a l
 imited number of model simulations and emulates the system responses. Ad
 ditionally\, to select these ‘training’ points\, an evidence-based activ
 e learning procedure is incorporated to adaptively refine the posterior 
 distribution of the parameters. The proposed approach allows the estimat
 ion of the posterior distribution of both the parameters of interest of 
 the model\, and the standard deviation of the observational noise\, allo
 wing a complete characterization of the model and the observational unce
 rtainty. The proposed methodology is applied to an illustrative example\
 , demonstrating its efficiency in estimating the posterior distribution 
 with a low computational cost. This framework is general\, applicable to
  a variety of problems involving complex or computationally expensive mo
 dels.
URL:https://eurodyn2026.org/submission/152
END:VEVENT
END:VCALENDAR