Solution of physics-based inverse problems using conditional generative adversarial networks with full gradient penalty

DOI10.1016/j.cma.2023.116338arXiv2306.04895MaRDI QIDQ6147033

Javier Murgoitio-Esandi, Assad A. Oberai, Deep Ray, Agnimitra Dasgupta

Publication date: 15 January 2024

Published in: Computer Methods in Applied Mechanics and Engineering (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2306.04895

zbMATH Keywords

Bayesian inferenceuncertainty quantificationgenerative modelsdeep learningconditional generative adversarial networksPDE-based inverse problems

Mathematics Subject Classification ID

Statistics (62-XX) Geophysics (86-XX)

Cites Work

• Unnamed Item
• Unnamed Item
• Weak convergence and optimal scaling of random walk Metropolis algorithms
• Solution of physics-based Bayesian inverse problems with deep generative priors
• Dimension-independent likelihood-informed MCMC
• Solution of the time-harmonic viscoelastic inverse problem with interior data in two dimensions
• Handbook of Markov Chain Monte Carlo
• Shear wave speed recovery in transient elastography and supersonic imaging using propagating fronts
• Monte Carlo strategies in scientific computing
• Priors in Bayesian Deep Learning: A Review