Subspace Accelerated Measure Transport Methods for Fast and Scalable Sequential Experimental Design, with Application to Photoacoustic Imaging
SIAM/ASA Journal on Uncertainty Quantification · 31.12.2026 · 14 (4) · 1476-1503
DOI: 10.1137/25m1735329 ↗Kurz gefasst
Fachartikel in der Zeitschrift SIAM/ASA Journal on Uncertainty Quantification, online erschienen am 15.10.2026. Verfasst von Tiangang Cui und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 60 Literaturangaben. Gefördert von: Australian Research Council, Carl-Zeiss-Stiftung.
Abstract (Originalsprache)
We propose a novel approach for sequential optimal experimental design (sOED) for Bayesian inverse problems involving expensive models with high-dimensional unknown parameters. This work focuses on designs that maximize the expected information gain (EIG) from prior to posterior, a task that is computationally very challenging in non-Gaussian settings. This challenge is amplified in sOED, as the incremental EIG (iEIG) must be repeatedly approximated across distinct stages, with both prior and posterior distributions being intractable. To address this, we derive a general-purpose, derivative-based upper bound for the iEIG, which not only guides design placement but also enables the construction of projectors onto likelihood-informed subspaces, facilitating parameter dimension reduction. By combining this approach with conditional measure transport maps for the sequence of posteriors, we develop a unified sOED and amortized inference framework scalable to high- and infinite-dimensional problems. Numerical experiments for two inverse problems governed by PDEs demonstrate the effectiveness of designs by maximizing the proposed bound.
Förderung
- Australian Research Council
- Carl-Zeiss-Stiftung
Zitieren
APA
Cui, T., Koval, K., Herzog, R., & Scheichl, R. (2026). Subspace Accelerated Measure Transport Methods for Fast and Scalable Sequential Experimental Design, with Application to Photoacoustic Imaging. SIAM/ASA Journal on Uncertainty Quantification, 14(4), 1476-1503. https://doi.org/10.1137/25m1735329
Vancouver
Cui T, Koval K, Herzog R, Scheichl R. Subspace Accelerated Measure Transport Methods for Fast and Scalable Sequential Experimental Design, with Application to Photoacoustic Imaging. SIAM/ASA Journal on Uncertainty Quantification. 2026;14(4):1476-1503. doi:10.1137/25m1735329
BibTeX
@article{Cui2026,
author = {Cui, Tiangang and Koval, Karina and Herzog, Roland and Scheichl, Robert},
title = {Subspace Accelerated Measure Transport Methods for Fast and Scalable Sequential Experimental Design, with Application to Photoacoustic Imaging},
journal = {SIAM/ASA Journal on Uncertainty Quantification},
year = {2026},
volume = {14},
number = {4},
pages = {1476--1503},
doi = {10.1137/25m1735329}
}
EUVIMED