Multilevel dimension-independent likelihood-informed MCMC for large-scale inverse problems

We present a non-trivial integration of dimension-independent likelihood-informed (DILI) MCMC (Cui et al 2016) and the multilevel MCMC (Dodwell et al 2015) to explore the hierarchy of posterior distributions. This integration offers several advantages: First, DILI-MCMC employs an intrinsic likelihoo...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cui, Tiangang (VerfasserIn) , Detommaso, Gianluca (VerfasserIn) , Scheichl, Robert (VerfasserIn)
Dokumenttyp: Article (Journal)
Sprache:Englisch
Veröffentlicht: 5 February 2024
In: Inverse problems
Year: 2024, Jahrgang: 40, Heft: 3, Pages: 1-33
ISSN:1361-6420
DOI:10.1088/1361-6420/ad1e2c
Online-Zugang:Verlag, kostenfrei, Volltext: https://doi.org/10.1088/1361-6420/ad1e2c
Verlag, kostenfrei, Volltext: https://dx.doi.org/10.1088/1361-6420/ad1e2c
Volltext
Verfasserangaben:Tiangang Cui, Gianluca Detommaso and Robert Scheichl

MARC

LEADER 00000caa a2200000 c 4500
001 1891362062
003 DE-627
005 20241205142536.0
007 cr uuu---uuuuu
008 240617s2024 xx |||||o 00| ||eng c
024 7 |a 10.1088/1361-6420/ad1e2c  |2 doi 
035 |a (DE-627)1891362062 
035 |a (DE-599)KXP1891362062 
035 |a (OCoLC)1475299855 
040 |a DE-627  |b ger  |c DE-627  |e rda 
041 |a eng 
084 |a 27  |2 sdnb 
100 1 |a Cui, Tiangang  |e VerfasserIn  |0 (DE-588)1333250436  |0 (DE-627)1891362518  |4 aut 
245 1 0 |a Multilevel dimension-independent likelihood-informed MCMC for large-scale inverse problems  |c Tiangang Cui, Gianluca Detommaso and Robert Scheichl 
264 1 |c 5 February 2024 
300 |b Illustrationen 
300 |a 33 
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 17.06.2024 
520 |a We present a non-trivial integration of dimension-independent likelihood-informed (DILI) MCMC (Cui et al 2016) and the multilevel MCMC (Dodwell et al 2015) to explore the hierarchy of posterior distributions. This integration offers several advantages: First, DILI-MCMC employs an intrinsic likelihood-informed subspace (LIS) (Cui et al 2014)—which involves a number of forward and adjoint model simulations—to design accelerated operator-weighted proposals. By exploiting the multilevel structure of the discretised parameters and discretised forward models, we design a Rayleigh-Ritz procedure to significantly reduce the computational effort in building the LIS and operating with DILI proposals. Second, the resulting DILI-MCMC can drastically improve the sampling efficiency of MCMC at each level, and hence reduce the integration error of the multilevel algorithm for fixed CPU time. Numerical results confirm the improved computational efficiency of the multilevel DILI approach. 
700 1 |a Detommaso, Gianluca  |e VerfasserIn  |0 (DE-588)1189080567  |0 (DE-627)1667811894  |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  |t Inverse problems  |d Bristol [u.a.] : Inst., 1985  |g 40(2024), 3, Artikel-ID 035005, Seite 1-33  |h Online-Ressource  |w (DE-627)270426108  |w (DE-600)1477292-9  |w (DE-576)078189306  |x 1361-6420  |7 nnas  |a Multilevel dimension-independent likelihood-informed MCMC for large-scale inverse problems 
773 1 8 |g volume:40  |g year:2024  |g number:3  |g elocationid:035005  |g pages:1-33  |g extent:33  |a Multilevel dimension-independent likelihood-informed MCMC for large-scale inverse problems 
856 4 0 |u https://doi.org/10.1088/1361-6420/ad1e2c  |x Verlag  |x Resolving-System  |z kostenfrei  |3 Volltext 
856 4 0 |u https://dx.doi.org/10.1088/1361-6420/ad1e2c  |x Verlag  |z kostenfrei  |3 Volltext 
951 |a AR 
992 |a 20240617 
993 |a Article 
994 |a 2024 
998 |g 1173753842  |a Scheichl, Robert  |m 1173753842:Scheichl, Robert  |d 110000  |d 110400  |e 110000PS1173753842  |e 110400PS1173753842  |k 0/110000/  |k 1/110000/110400/  |p 3  |y j 
999 |a KXP-PPN1891362062  |e 4539508725 
BIB |a Y 
SER |a journal 
JSO |a {"id":{"eki":["1891362062"],"doi":["10.1088/1361-6420/ad1e2c"]},"type":{"media":"Online-Ressource","bibl":"article-journal"},"name":{"displayForm":["Tiangang Cui, Gianluca Detommaso and Robert Scheichl"]},"language":["eng"],"recId":"1891362062","title":[{"title_sort":"Multilevel dimension-independent likelihood-informed MCMC for large-scale inverse problems","title":"Multilevel dimension-independent likelihood-informed MCMC for large-scale inverse problems"}],"person":[{"given":"Tiangang","family":"Cui","role":"aut","display":"Cui, Tiangang"},{"role":"aut","display":"Detommaso, Gianluca","family":"Detommaso","given":"Gianluca"},{"display":"Scheichl, Robert","role":"aut","given":"Robert","family":"Scheichl"}],"relHost":[{"note":["Gesehen am 21.09.2020"],"origin":[{"dateIssuedKey":"1985","publisherPlace":"Bristol [u.a.]","dateIssuedDisp":"1985-","publisher":"Inst."}],"physDesc":[{"extent":"Online-Ressource"}],"recId":"270426108","name":{"displayForm":["Institute of Physics, London and Bristol"]},"pubHistory":["1.1985 -"],"id":{"zdb":["1477292-9"],"eki":["270426108"],"issn":["1361-6420"]},"title":[{"title":"Inverse problems","title_sort":"Inverse problems","subtitle":"an international journal on the theory and practice of inverse problems, inverse methods and computerized inversion of data"}],"part":{"extent":"33","year":"2024","volume":"40","issue":"3","pages":"1-33","text":"40(2024), 3, Artikel-ID 035005, Seite 1-33"},"titleAlt":[{"title":"an international journal of inverse problems, inverse methods and computerized inversion of data"},{"title":"a journal recognized by the European Physical Society"}],"disp":"Multilevel dimension-independent likelihood-informed MCMC for large-scale inverse problemsInverse problems","type":{"media":"Online-Ressource","bibl":"periodical"},"language":["eng"]}],"physDesc":[{"noteIll":"Illustrationen","extent":"33 S."}],"note":["Gesehen am 17.06.2024"],"origin":[{"dateIssuedDisp":"5 February 2024","dateIssuedKey":"2024"}]} 
SRT |a CUITIANGANMULTILEVEL5202