Deep generative models in inversion: The impact of the generator's nonlinearity and development of a new approach based on a variational autoencoder

When solving inverse problems in geophysical imaging, deep generative models (DGMs) may be used to enforce the solution to display highly structured spatial patterns which are supported by independent information (e.g. the geological setting) of the subsurface. In such case, inversion may be formula...

Full description

Saved in:

Bibliographic Details
Published in:	Computers & geosciences Vol. 152; p. 104762
Main Authors:	Lopez-Alvis, Jorge, Laloy, Eric, Nguyen, Frédéric, Hermans, Thomas
Format:	Journal Article
Language:	English
Published:	Elsevier Ltd 01.07.2021
Subjects:	Computers in Earth Sciences Conceptual frameworks Deep generative model Deep learning Engineering, computing & technology Generative functions Geological prior information Geological setting Geological, petroleum & mining engineering Geophysical imaging Geophysical inversion geophysics Gradient-based method Géologie, ingénierie du pétrole & des mines Information Systems Ingénierie, informatique & technologie Markov chain Markov Chain Monte-Carlo Stochastic gradient descent tomography topology Training parameters Variational autoencoder Deep learning Variational autoencoder Deep generative model Geophysical inversion Geological prior information Stochastic gradient descent
ISSN:	0098-3004, 1873-7803, 1873-7803
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Be the first to leave a comment!