Solving inverse problems using conditional invertible neural networks

Abstract

Inverse modeling for computing a high-dimensional spatially-varying property field from indirect sparse and noisy observations is a challenging problem. This is due to the complex physical system of interest often expressed in the form of multiscale PDEs, the high-dimensionality of the spatial property of interest, and the incomplete and noisy nature of observations. To address these challenges, we develop a model that maps the given observations to the unknown input field in the form of a surrogate model. This inverse surrogate model will then allow us to estimate the unknown input field for any given sparse and noisy output observations. Here, the inverse mapping is limited to a broad prior distribution of the input field with which the surrogate model is trained. In this work, we construct a two- and three-dimensional inverse surrogate models consisting of an invertible and a conditional neural network trained in an end-to-end fashion with limited training data. The invertible network is developed using a flow-based generative model. The developed inverse surrogate model is then applied for an inversion task of a multiphase flow problem where given the pressure and saturation observations the aim is to recover a high-dimensional non-Gaussian permeability field where the two facies consist of heterogeneous permeability and varying length-scales. For both the two- and three-dimensional surrogate models, the predicted sample realizations of the non-Gaussian permeability field are diverse with the predictive mean being close to the ground truth even when the model is trained with limited data.

arXiv comment58 pages, 20 figures

References in corpus (29)

• Adam: A Method for Stochastic Optimization
• Conditional Generative Adversarial Nets
• Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting
• Deep Convolutional Neural Network for Inverse Problems in Imaging
• NICE: Non-linear Independent Components Estimation
• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• Bayesian Deep Convolutional Encoder-Decoder Networks for Surrogate Modeling and Uncertainty Quantification
• Density estimation using Real NVP
• Deep Learning Methods for Reynolds-Averaged Navier-Stokes Simulations of Airfoil Flows
• Reconstruction of three-dimensional porous media using generative adversarial neural networks
• Solving ill-posed inverse problems using iterative deep neural networks
• Deep UQ: Learning deep neural network surrogate models for high dimensional uncertainty quantification
• The History Began from AlexNet: A Comprehensive Survey on Deep Learning Approaches
• Training-image based geostatistical inversion using a spatial generative adversarial neural network
• Deep D-bar: Real time Electrical Impedance Tomography Imaging with Deep Neural Networks
• Deep convolutional encoder-decoder networks for uncertainty quantification of dynamic multiphase flow in heterogeneous media
• Modeling the Dynamics of PDE Systems with Physics-Constrained Deep Auto-Regressive Networks
• Deep autoregressive neural networks for high-dimensional inverse problems in groundwater contaminant source identification
• Inversion using a new low-dimensional representation of complex binary geological media based on a deep neural network
• Guided Image Generation with Conditional Invertible Neural Networks
• Integration of adversarial autoencoders with residual dense convolutional networks for estimation of non-Gaussian hydraulic conductivities
• Analyzing Inverse Problems with Invertible Neural Networks
• Quantifying model form uncertainty in Reynolds-averaged turbulence models with Bayesian deep neural networks
• Deep Generative Adversarial Networks for Compressed Sensing Automates MRI
• An iterative local updating ensemble smoother for estimation and uncertainty assessment of hydrologic model parameters with multimodal distributions
• Solving Electrical Impedance Tomography with Deep Learning
• Parametrization and generation of geological models with generative adversarial networks
• Multi-fidelity Generative Deep Learning Turbulent Flows
• Reversible Recurrent Neural Networks

Cited by in corpus (15)

• Normalizing field flows: Solving forward and inverse stochastic differential equations using physics-informed flow models
• Bayesian multiscale deep generative model for the solution of high-dimensional inverse problems
• Conditional score-based diffusion models for solving inverse problems in mechanics
• Bi-fidelity Variational Auto-encoder for Uncertainty Quantification
• Variational Bayesian Optimal Experimental Design with Normalizing Flows
• Conditional Injective Flows for Bayesian Imaging
• Learning to solve Bayesian inverse problems: An amortized variational inference approach using Gaussian and Flow guides
• Variational Sequential Optimal Experimental Design using Reinforcement Learning
• Study of the asteroid Bennu using geodesyANNs and Osiris-Rex data
• Geodesy of irregular small bodies via neural density fields: geodesyNets
• Inverse Aerodynamic Design of Gas Turbine Blades using Probabilistic Machine Learning
• Reconstructing the Magnetic Field in an Arbitrary Domain via Data-driven Bayesian Methods and Numerical Simulations
• Invertible Surrogate Models: Joint surrogate modelling and reconstruction of Laser-Wakefield Acceleration by invertible neural networks
• Uncertainty quantification for ptychography using normalizing flows
• DAWN-FM: Data-Aware and Noise-Informed Flow Matching for Solving Inverse Problems