Non-intrusive optimal experimental design for large-scale nonlinear Bayesian inverse problems using a Bayesian approximation error approach
We consider optimal experimental design (OED) for nonlinear inverse problems within the Bayesian framework. Optimizing the data acquisition process for large-scale nonlinear Bayesian inverse problems is a computationally challenging task since the posterior is typically intractable and commonly-enco...
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
1 August 2025
|
| In: |
Journal of scientific computing
Year: 2025, Jahrgang: 104, Heft: 3, Pages: 1-28 |
| ISSN: | 1573-7691 |
| DOI: | 10.1007/s10915-025-03008-7 |
| Online-Zugang: | Verlag, kostenfrei, Volltext: https://doi.org/10.1007/s10915-025-03008-7 Verlag, kostenfrei, Volltext: https://link.springer.com/article/10.1007/s10915-025-03008-7 
|
| Verfasserangaben: | Karina Koval, Ruanui Nicholson |
MARC
| LEADER | 00000naa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 194375215X | ||
| 003 | DE-627 | ||
| 005 | 20251203113619.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 251203s2025 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1007/s10915-025-03008-7 |2 doi | |
| 035 | |a (DE-627)194375215X | ||
| 035 | |a (DE-599)KXP194375215X | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 28 |2 sdnb | ||
| 100 | 1 | |a Koval, Karina |e VerfasserIn |0 (DE-588)1360816879 |0 (DE-627)1920241647 |4 aut | |
| 245 | 1 | 0 | |a Non-intrusive optimal experimental design for large-scale nonlinear Bayesian inverse problems using a Bayesian approximation error approach |c Karina Koval, Ruanui Nicholson |
| 264 | 1 | |c 1 August 2025 | |
| 300 | |b Illustrationen | ||
| 300 | |a 28 | ||
| 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 03.12.2025 | ||
| 520 | |a We consider optimal experimental design (OED) for nonlinear inverse problems within the Bayesian framework. Optimizing the data acquisition process for large-scale nonlinear Bayesian inverse problems is a computationally challenging task since the posterior is typically intractable and commonly-encountered optimality criteria depend on the observed data. Since these challenges are not present in OED for linear Bayesian inverse problems, we propose an approach based on first linearizing the associated forward problem and then optimizing the experimental design. Replacing an accurate but costly model with some linear surrogate, while justified for certain problems, can lead to incorrect posteriors and sub-optimal designs if model discrepancy is ignored. To avoid this, we use the Bayesian approximation error (BAE) approach to formulate an A-optimal design objective for sensor selection that is aware of the model error. In line with recent developments, we prove that this uncertainty-aware objective is independent of the exact choice of linearization. This key observation facilitates the formulation of an uncertainty-aware OED objective function using a completely trivial linear map, the zero map, as a surrogate to the forward dynamics. The result is also extended to marginalized OED problems, accommodating uncertainties arising from both linear approximations and unknown auxiliary parameters. Our approach only requires parameter and data sample pairs, hence it is particularly well-suited for black box forward models. We demonstrate the effectiveness of our method for finding optimal designs in an idealized subsurface flow inverse problem and for tsunami detection. | ||
| 650 | 4 | |a 62-08 | |
| 650 | 4 | |a 62E17 | |
| 650 | 4 | |a 62F15 | |
| 650 | 4 | |a 65K05 | |
| 650 | 4 | |a Bayesian approximation error | |
| 650 | 4 | |a Bayesian inverse problems | |
| 650 | 4 | |a Linearization | |
| 650 | 4 | |a Optimal experimental design | |
| 700 | 1 | |a Nicholson, Ruanui |e VerfasserIn |0 (DE-588)1383284792 |0 (DE-627)1943752729 |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Journal of scientific computing |d New York, NY [u.a.] : Springer Science + Business Media B.V., 1986 |g 104(2025), 3 vom: Sept., Artikel-ID 98, Seite 1-28 |h Online-Ressource |w (DE-627)317878395 |w (DE-600)2017260-6 |w (DE-576)121466221 |x 1573-7691 |7 nnas |a Non-intrusive optimal experimental design for large-scale nonlinear Bayesian inverse problems using a Bayesian approximation error approach |
| 773 | 1 | 8 | |g volume:104 |g year:2025 |g number:3 |g month:09 |g elocationid:98 |g pages:1-28 |g extent:28 |a Non-intrusive optimal experimental design for large-scale nonlinear Bayesian inverse problems using a Bayesian approximation error approach |
| 856 | 4 | 0 | |u https://doi.org/10.1007/s10915-025-03008-7 |x Verlag |x Resolving-System |z kostenfrei |3 Volltext |
| 856 | 4 | 0 | |u https://link.springer.com/article/10.1007/s10915-025-03008-7 |x Verlag |z kostenfrei |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20251203 | ||
| 993 | |a Article | ||
| 994 | |a 2025 | ||
| 998 | |g 1360816879 |a Koval, Karina |m 1360816879:Koval, Karina |d 700000 |d 708000 |e 700000PK1360816879 |e 708000PK1360816879 |k 0/700000/ |k 1/700000/708000/ |p 1 |x j | ||
| 999 | |a KXP-PPN194375215X |e 4819891707 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"physDesc":[{"extent":"28 S.","noteIll":"Illustrationen"}],"relHost":[{"origin":[{"dateIssuedDisp":"1986-","publisher":"Springer Science + Business Media B.V. ; Kluwer","dateIssuedKey":"1986","publisherPlace":"New York, NY [u.a.] ; London [u.a.]"}],"id":{"issn":["1573-7691"],"eki":["317878395"],"zdb":["2017260-6"]},"physDesc":[{"extent":"Online-Ressource"}],"title":[{"title_sort":"Journal of scientific computing","title":"Journal of scientific computing"}],"type":{"bibl":"periodical","media":"Online-Ressource"},"note":["Gesehen am 01.11.05"],"disp":"Non-intrusive optimal experimental design for large-scale nonlinear Bayesian inverse problems using a Bayesian approximation error approachJournal of scientific computing","language":["eng"],"recId":"317878395","pubHistory":["1.1986 -"],"part":{"text":"104(2025), 3 vom: Sept., Artikel-ID 98, Seite 1-28","volume":"104","extent":"28","year":"2025","issue":"3","pages":"1-28"}}],"name":{"displayForm":["Karina Koval, Ruanui Nicholson"]},"origin":[{"dateIssuedDisp":"1 August 2025","dateIssuedKey":"2025"}],"id":{"eki":["194375215X"],"doi":["10.1007/s10915-025-03008-7"]},"type":{"media":"Online-Ressource","bibl":"article-journal"},"note":["Gesehen am 03.12.2025"],"recId":"194375215X","language":["eng"],"person":[{"role":"aut","display":"Koval, Karina","roleDisplay":"VerfasserIn","given":"Karina","family":"Koval"},{"role":"aut","roleDisplay":"VerfasserIn","display":"Nicholson, Ruanui","given":"Ruanui","family":"Nicholson"}],"title":[{"title_sort":"Non-intrusive optimal experimental design for large-scale nonlinear Bayesian inverse problems using a Bayesian approximation error approach","title":"Non-intrusive optimal experimental design for large-scale nonlinear Bayesian inverse problems using a Bayesian approximation error approach"}]} | ||
| SRT | |a KOVALKARINNONINTRUSI1202 | ||