000014819 001__ 14819
000014819 005__ 20161115100203.0
000014819 04107 $$aeng
000014819 046__ $$k2016-08-21
000014819 100__ $$aGhattas, Omar
000014819 24500 $$aBig data meets big models: Towards exascale Bayesian inverse problems (INVITED)

000014819 24630 $$n24.$$p24th International Congress of Theoretical and Applied Mechanics - Book of Papers
000014819 260__ $$bInternational Union of Theoretical and Applied Mechanics, 2016
000014819 506__ $$arestricted
000014819 520__ $$2eng$$aPredictive models of complex systems often contain numerous uncertain parameters. Rapidly expanding volumes of observational data present opportunities to reduce these uncertainties via solution of inverse problems. Bayesian inference provides a systematic framework for inferring model parameters with associated uncertainties from (possibly noisy) data and any prior information. However, solution of the Bayesian inverse problem via conventional Markov chain Monte Carlo methods remains prohibitive for expensive models and high-dimensional parameters. Despite the large size of observational datasets, typically they provide only sparse information on model parameters. Based on this property we design MCMC methods that adapt to the structure of the posterior probability and exploit an effectively-reduced parameter dimension. We apply the methodology to an inverse problem for Antarctic ice sheet flow.

000014819 540__ $$aText je chráněný podle autorského zákona č. 121/2000 Sb.
000014819 653__ $$a

000014819 7112_ $$a24th International Congress of Theoretical and Applied Mechanics$$cMontreal (CA)$$d2016-08-21 / 2016-08-26$$gICTAM2016
000014819 720__ $$aGhattas, Omar
000014819 8560_ $$ffischerc@itam.cas.cz
000014819 8564_ $$s44060$$uhttps://invenio.itam.cas.cz/record/14819/files/TS.FS02-1.01.pdf$$yOriginal version of the author's contribution as presented on CD,  page 3094, code TS.FS02-1.01
.
000014819 962__ $$r13812
000014819 980__ $$aPAPER