The efficacy and generalizability of conditional GANs for posterior inference in physics-based inverse problems
Preprint
Abstract
In this work, we train conditional Wasserstein generative adversarial networks to effectively sample from the posterior of physics-based Bayesian inference problems. The generator is constructed using a U-Net architecture, with the latent information injected using conditional instance normalization. The former facilitates a multiscale inverse map, while the latter enables the decoupling of the latent space dimension from the dimension of the measurement, and introduces stochasticity at all scales of the U-Net. We solve PDE-based inverse problems to demonstrate the performance of our approach in quantifying the uncertainty in the inferred field. Further, we show the generator can learn inverse maps which are local in nature, which in turn promotes generalizability when testing with out-of-distribution samples.Keywords
All Related Versions
This publication has 17 references indexed in Scilit:
- Solution of physics-based Bayesian inverse problems with deep generative priorsComputer Methods in Applied Mechanics and Engineering, 2022
- Deep Learning Techniques for Inverse Problems in ImagingIEEE Journal on Selected Areas in Information Theory, 2020
- Validation of Quantitative Linear and Nonlinear Compression Elastography2018
- Solving ill-posed inverse problems using iterative deep neural networksInverse Problems, 2017
- The Bayesian Approach to Inverse Problems2017
- A nonlinear elasticity phantom containing spherical inclusionsPhysics in Medicine & Biology, 2012
- A Review of the Mathematical and Computational Foundations of Biomechanical Imaging2009
- Elastic modulus imaging: some exact solutions of the compressible elastography inverse problemPhysics in Medicine & Biology, 2007
- Inverse problems in atmospheric science and their applicationJournal of Physics: Conference Series, 2005
- Resolution of seismic waveform inversion: Bayes versus OccamInverse Problems, 1997