Multilevel delayed acceptance MCMC

.We develop a fast and scalable computational framework to solve Bayesian optimal experimental design problems governed by partial differential equations (PDEs) with application to optimal sensor placement by maximizing expected information gain (EIG). Such problems are particularly challenging due...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Lykkegaard, Mikkel (VerfasserIn) , Dodwell, T. J. (VerfasserIn) , Fox, C. (VerfasserIn) , Mingas, G. (VerfasserIn) , Scheichl, Robert (VerfasserIn)
Dokumenttyp: Article (Journal)
Sprache:Englisch
Veröffentlicht: January 25, 2023
In: SIAM ASA journal on uncertainty quantification
Year: 2023, Jahrgang: 11, Heft: 1, Pages: 1-30
ISSN:2166-2525
DOI:10.1137/22M1476770
Online-Zugang:Verlag, lizenzpflichtig, Volltext: https://doi.org/10.1137/22M1476770
Verlag, lizenzpflichtig, Volltext: https://epubs.siam.org/doi/10.1137/22M1476770
Volltext
Verfasserangaben:M.B. Lykkegaard, T.J. Dodwell, C. Fox, G. Mingas, and R. Scheichl

MARC

LEADER 00000caa a2200000 c 4500
001 1845610938
003 DE-627
005 20230706210257.0
007 cr uuu---uuuuu
008 230517s2023 xx |||||o 00| ||eng c
024 7 |a 10.1137/22M1476770  |2 doi 
035 |a (DE-627)1845610938 
035 |a (DE-599)KXP1845610938 
035 |a (OCoLC)1389530150 
040 |a DE-627  |b ger  |c DE-627  |e rda 
041 |a eng 
084 |a 27  |2 sdnb 
100 1 |a Lykkegaard, Mikkel  |e VerfasserIn  |0 (DE-588)1289652783  |0 (DE-627)1845611586  |4 aut 
245 1 0 |a Multilevel delayed acceptance MCMC  |c M.B. Lykkegaard, T.J. Dodwell, C. Fox, G. Mingas, and R. Scheichl 
264 1 |c January 25, 2023 
300 |a 30 
336 |a Text  |b txt  |2 rdacontent 
337 |a Computermedien  |b c  |2 rdamedia 
338 |a Online-Ressource  |b cr  |2 rdacarrier 
500 |a Gesehen am 17.05.2023 
520 |a .We develop a fast and scalable computational framework to solve Bayesian optimal experimental design problems governed by partial differential equations (PDEs) with application to optimal sensor placement by maximizing expected information gain (EIG). Such problems are particularly challenging due to the curse of dimensionality for high-dimensional parameters and the expensive solution of large-scale PDEs. To address these challenges, we exploit two fundamental properties: (1) the low-rank structure of the Jacobian of the parameter-to-observable map, to extract the intrinsically low-dimensional data-informed subspace, and (2) a series of approximations of the EIG that reduce the number of PDE solves while retaining high correlation with the true EIG. Based on these properties, we propose an efficient offline-online decomposition for the optimization problem. The offline stage dominates the cost and entails precomputing all components that require PDE solves. The online stage optimizes sensor placement and does not require any PDE solves. For the online stage, we propose a new greedy algorithm that first places an initial set of sensors using leverage scores and then swaps the selected sensors with other candidates until certain convergence criteria are met, which we call a swapping greedy algorithm. We demonstrate the efficiency and scalability of the proposed method by both linear and nonlinear inverse problems. In particular, we show that the number of required PDE solves is small, independent of the parameter dimension, and only weakly dependent on the data dimension for both problems. 
700 1 |a Dodwell, T. J.  |e VerfasserIn  |4 aut 
700 1 |a Fox, C.  |e VerfasserIn  |4 aut 
700 1 |a Mingas, G.  |e VerfasserIn  |4 aut 
700 1 |a Scheichl, Robert  |d 1972-  |e VerfasserIn  |0 (DE-588)1173753842  |0 (DE-627)1043602305  |0 (DE-576)515668532  |4 aut 
773 0 8 |i Enthalten in  |a Society for Industrial and Applied Mathematics  |t SIAM ASA journal on uncertainty quantification  |d Philadelphia, Pa. : SIAM, 2013  |g 11(2023), 1, Seite 1-30  |h Online-Ressource  |w (DE-627)729754006  |w (DE-600)2690603-X  |w (DE-576)391400703  |x 2166-2525  |7 nnas 
773 1 8 |g volume:11  |g year:2023  |g number:1  |g pages:1-30  |g extent:30  |a Multilevel delayed acceptance MCMC 
856 4 0 |u https://doi.org/10.1137/22M1476770  |x Verlag  |x Resolving-System  |z lizenzpflichtig  |3 Volltext 
856 4 0 |u https://epubs.siam.org/doi/10.1137/22M1476770  |x Verlag  |z lizenzpflichtig  |3 Volltext 
951 |a AR 
992 |a 20230517 
993 |a Article 
994 |a 2023 
998 |g 1173753842  |a Scheichl, Robert  |m 1173753842:Scheichl, Robert  |d 110000  |d 110200  |d 110000  |d 110400  |d 700000  |d 728500  |e 110000PS1173753842  |e 110200PS1173753842  |e 110000PS1173753842  |e 110400PS1173753842  |e 700000PS1173753842  |e 728500PS1173753842  |k 0/110000/  |k 1/110000/110200/  |k 0/110000/  |k 1/110000/110400/  |k 0/700000/  |k 1/700000/728500/  |p 5  |y j 
999 |a KXP-PPN1845610938  |e 4323058934 
BIB |a Y 
SER |a journal 
JSO |a {"type":{"media":"Online-Ressource","bibl":"article-journal"},"name":{"displayForm":["M.B. Lykkegaard, T.J. Dodwell, C. Fox, G. Mingas, and R. Scheichl"]},"language":["eng"],"recId":"1845610938","id":{"doi":["10.1137/22M1476770"],"eki":["1845610938"]},"note":["Gesehen am 17.05.2023"],"origin":[{"dateIssuedKey":"2023","dateIssuedDisp":"January 25, 2023"}],"person":[{"family":"Lykkegaard","given":"Mikkel","role":"aut","display":"Lykkegaard, Mikkel"},{"role":"aut","display":"Dodwell, T. J.","given":"T. J.","family":"Dodwell"},{"display":"Fox, C.","role":"aut","given":"C.","family":"Fox"},{"given":"G.","family":"Mingas","display":"Mingas, G.","role":"aut"},{"family":"Scheichl","given":"Robert","role":"aut","display":"Scheichl, Robert"}],"title":[{"title":"Multilevel delayed acceptance MCMC","title_sort":"Multilevel delayed acceptance MCMC"}],"relHost":[{"name":{"displayForm":["Society for Industrial and Applied Mathematics ; American Statistical Association"]},"recId":"729754006","id":{"issn":["2166-2525"],"eki":["729754006"],"zdb":["2690603-X"]},"pubHistory":["1.2013 -"],"note":["Gesehen am 21.05.13"],"origin":[{"publisherPlace":"Philadelphia, Pa.","dateIssuedKey":"2013","dateIssuedDisp":"2013-","publisher":"SIAM"}],"physDesc":[{"extent":"Online-Ressource"}],"language":["eng"],"disp":"Society for Industrial and Applied MathematicsSIAM ASA journal on uncertainty quantification","type":{"bibl":"periodical","media":"Online-Ressource"},"titleAlt":[{"title":"JUQ"},{"title":"Journal on uncertainty quantification"}],"title":[{"title_sort":"SIAM ASA journal on uncertainty quantification","title":"SIAM ASA journal on uncertainty quantification"}],"corporate":[{"display":"Society for Industrial and Applied Mathematics","role":"aut"},{"role":"isb","display":"American Statistical Association"}],"part":{"pages":"1-30","text":"11(2023), 1, Seite 1-30","issue":"1","year":"2023","volume":"11","extent":"30"}}],"physDesc":[{"extent":"30 S."}]} 
SRT |a LYKKEGAARDMULTILEVEL2520