A dimension-reduced variational approach for solving physics-based inverse problems using generative adversarial network priors and normalizing flows
- 20 December 2023
- journal article
- research article
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 420
Abstract
No abstract availableKeywords
All Related Versions
Funding Information
- University of Southern California
- National Science Foundation
- US Department of the Army
- National Science Foundation Division of Civil Mechanical and Manufacturing Innovation (CMMI 16-63667)
- Army Research Office (W911NF2010050)
This publication has 34 references indexed in Scilit:
- REIN: Reliability Estimation via Importance sampling with Normalizing flowsReliability Engineering & System Safety, 2023
- Solution of physics-based inverse problems using conditional generative adversarial networks with full gradient penaltyComputer Methods in Applied Mechanics and Engineering, 2023
- Solution of physics-based Bayesian inverse problems with deep generative priorsComputer Methods in Applied Mechanics and Engineering, 2022
- Variational inference with NoFAS: Normalizing flow with adaptive surrogate for computationally expensive modelsJournal of Computational Physics, 2022
- Learning functional priors and posteriors from data and physicsJournal of Computational Physics, 2022
- An efficient optimization based microstructure reconstruction approach with multiple loss functionsComputational Materials Science, 2021
- Generative adversarial network in medical imaging: A reviewMedical Image Analysis, 2019
- Inverse problems: From regularization to Bayesian inferenceWIREs Computational Statistics, 2018
- Bayesian inference with optimal mapsJournal of Computational Physics, 2012
- Phase retrievalCommunications in Mathematical Physics, 1984