BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20220812T074335Z
LOCATION:Singapore Room
DTSTART;TZID=Europe/Stockholm:20220628T180000
DTEND;TZID=Europe/Stockholm:20220628T183000
UID:submissions.pasc-conference.org_PASC22_sess149_msa258@linklings.com
SUMMARY:Gradient-Accelerated Bayesian Inference for Non-Linear Uncertainty
  Quantification of High-Dimensional Models
DESCRIPTION:Minisymposium\n\nGradient-Accelerated Bayesian Inference for N
 on-Linear Uncertainty Quantification of High-Dimensional Models\n\nGebraad
 , Zunino, Fichtner\n\nWe discuss current state-of-the-art methods to updat
 e the knowledge on the Earth, integrating prior beliefs with various data 
 to answer questions about the subsurface. By integrating modern waveform m
 odelling techniques and gradient-based sampling we are able to push the bo
 unds of of geophysical knowledge in a quantifiable way.<br /><br />Specifi
 cally, we couple the Hamiltonian Monte Carlo (HMC) sampler to Full-Wavefor
 m inversion (FWI) techniques enabled by Salvus, a GPU accelerated HPC read
 y waveform modeller. Clasically, uncertainty in FWI is quantified only in 
 a limited sense, by e.g. the computation of point spread functions, Hessia
 n approximations or simply by assessing the reconstruction of a synthetic 
 dataset. These limitations were required by the computational cost of FWI 
 evaluations. However, using HMC and scalable waveform modelling, fully pro
 babilistic sampling becomes possible. By sampling the resulting inverse pr
 oblems in FWI in a fully non-linear fashion (i.e. by using HMC), uncertain
 ty is assessed with any assumptios. The main benefits of this is a high co
 nfidence in the resulting uncertainty quantification as well as a reduced 
 regularization requirement on ill-constrained parameters. This last point 
 is especially important for parameters such as density, for which it is un
 clear how they are constrained by seismological data in FWI settings.\n\nD
 omain: Climate, Weather and Earth Sciences, Computer Science and Applied M
 athematics, Physics
END:VEVENT
END:VCALENDAR