Uncertainty Quantification with Generative Models
Machine Learning
2019-10-24 v1 Cosmology and Nongalactic Astrophysics
Machine Learning
Abstract
We develop a generative model-based approach to Bayesian inverse problems, such as image reconstruction from noisy and incomplete images. Our framework addresses two common challenges of Bayesian reconstructions: 1) It makes use of complex, data-driven priors that comprise all available information about the uncorrupted data distribution. 2) It enables computationally tractable uncertainty quantification in the form of posterior analysis in latent and data space. The method is very efficient in that the generative model only has to be trained once on an uncorrupted data set, after that, the procedure can be used for arbitrary corruption types.
Cite
@article{arxiv.1910.10046,
title = {Uncertainty Quantification with Generative Models},
author = {Vanessa Böhm and François Lanusse and Uroš Seljak},
journal= {arXiv preprint arXiv:1910.10046},
year = {2019}
}
Comments
accepted submission to the Bayesian Deep Learning NeurIPS 2019 Workshop