Gradient-based deterministic inversion of geophysical data with Generative Adversarial Networks: is it feasible?

Abstract

Global probabilistic inversion within the latent space learned by a Generative Adversarial Network (GAN) has been recently demonstrated. Compared to inversion on the original model space, using the latent space of a trained GAN can offer the following benefits: (1) the generated model proposals are geostatistically consistent with the prescribed prior training image (TI), and (2) the parameter space is reduced by orders of magnitude compared to the original model space. Nevertheless, exploring the learned latent space by state-of-the-art Markov chain Monte Carlo (MCMC) methods may still require a large computational effort. As an alternative, parameters in this latent space could possibly be optimized with much less computationally expensive gradient-based methods. We show that due to the typically highly nonlinear relationship between the latent space and the associated output space of a GAN, gradient-based deterministic inversion may fail even when considering a linear forward physical model. We tested two deterministic inversion approaches: a quasi-Newton gradient descent using the Adam algorithm and a Gauss-Newton (GN) method that makes use of the Jacobian matrix calculated by finite-differencing. For a channelized binary TI and a synthetic linear crosshole ground penetrating radar (GPR) tomography problem involving 576 measurements with low noise, we observe that when allowing for a total of 10,000 iterations only 13% of the gradient descent trials locate a solution that has the required data misfit. The tested GN inversion was unable to recover a solution with the appropriate data misfit. Our results suggest that deterministic inversion performance strongly depends on the inversion approach, starting model, true reference model, number of iterations and noise realization. In contrast, computationally-expensive probabilistic global optimization always finds an appropriate solution.

References in corpus (14)

• Adam: A Method for Stochastic Optimization
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Generative Adversarial Networks
• Generative Adversarial Networks: An Overview
• Reconstruction of three-dimensional porous media using generative adversarial neural networks
• Training-image based geostatistical inversion using a spatial generative adversarial neural network
• Adaptive Gradient Methods with Dynamic Bound of Learning Rate
• Texture Synthesis with Spatial Generative Adversarial Networks
• Precise Recovery of Latent Vectors from Generative Adversarial Networks
• Metrics for Deep Generative Models
• Inverting The Generator Of A Generative Adversarial Network (II)
• The Riemannian Geometry of Deep Generative Models
• Inverting The Generator Of A Generative Adversarial Network
• Generative Adversarial Networks for Model Order Reduction in Seismic Full-Waveform Inversion

Cited by in corpus (10)

• Deep-learning-based surrogate flow modeling and geological parameterization for data assimilation in 3D subsurface flow
• Integration of adversarial autoencoders with residual dense convolutional networks for estimation of non-Gaussian hydraulic conductivities
• Deep generative models in inversion: a review and development of a new approach based on a variational autoencoder
• 3D CNN-PCA: A Deep-Learning-Based Parameterization for Complex Geomodels
• Deep Learning for Simultaneous Inference of Hydraulic and Transport Properties
• Bayesian multiscale deep generative model for the solution of high-dimensional inverse problems
• Continuous conditional generative adversarial networks for data-driven solutions of poroelasticity with heterogeneous material properties
• Adaptive sequential Monte Carlo for posterior inference and model selection among complex geological priors
• Data-Space Inversion Using a Recurrent Autoencoder for Time-Series Parameterization
• Diffusion models for multivariate subsurface generation and efficient probabilistic inversion