Conditional score-based generative models for solving physics-based inverse problems

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

Keywords: inverse problems, score-based models, conditional generative models, Bayesian inference

TL;DR: We use conditional score-based generative models to solve physics-based inverse problems and compare their performance against conditional GANs

Abstract: We propose to sample from high-dimensional posterior distributions arising in physics-based inverse problems using conditional score-based generative models. The proposed approach trains a noise-conditional score network to approximate the score function of the posterior distribution. Then, the network is used to sample from the posterior distribution through annealed Langevin dynamics. The proposed method is applicable even when we can only simulate the forward problem. We apply it to two physics-based inverse problems and compare its performance with conditional generative adversarial networks. Results show that conditional score-based generative models can reliably perform Bayesian inference.

Submission Number: 23