Abstract:ABSTRACTCasting a geophysical inverse problem into a Bayesian setting is often discouraged by the computational workload needed to run many forward modelling evaluations. Here we present probabilistic inversions of electrical resistivity tomography data in which the forward operator is replaced by a trained residual neural network that learns the non‐linear mapping between the resistivity model and the apparent resistivity values. The use of this specific architecture can provide some advantages over standard convolutional networks as it mitigates the vanishing gradient problem that might affect deep networks. The modelling error introduced by the network approximation is properly taken into account and propagated onto the estimated model uncertainties. One crucial aspect of any machine learning application is the definition of an appropriate training set. We draw the models forming the training and validation sets from previously defined prior distributions, while a finite element code provides the associated datasets. We apply the approach to two probabilistic inversion frameworks: A Markov chain Monte Carlo algorithm is applied to synthetic data, while an ensemble‐based algorithm is employed for the field measurements. For both the synthetic and field tests, the outcomes of the proposed method are benchmarked against the predictions obtained when the finite element code constitutes the forward operator. Our experiments illustrate that the network can effectively approximate the forward mapping even when a relatively small training set is created. The proposed strategy provides a forward operator that is three orders of magnitude faster than the accurate but computationally expensive finite element code. Our approach also yields the most likely solutions and uncertainty quantifications comparable to those estimated when the finite element modelling is employed. The presented method allows solving the Bayesian electrical resistivity tomography with a reasonable computational cost and limited hardware resources.

Fast Bayesian Inversion of Airborne Electromagnetic Data Based on the Invertible Neural Network

Convolutional Neural Network Inversion Of Airborne Transient Electromagnetic Data

Instantaneous inversion of airborne electromagnetic data based on deep learning

A Method for Inverting Shallow Sea Acoustic Parameters Based on the Backward Feedback Neural Network Model

Physics-guided Deep Learning-Based Inversion for Airborne Electromagnetic Data

Rapid Surrogate Modeling of Electromagnetic Data in Frequency Domain Using Neural Operator.

Inverting airborne electromagnetic data with machine learning

Probabilistic inversions of electrical resistivity tomography data with a machine learning‐based forward operator

Inversion of Low- to Medium-Frequency Velocities and Densities from AVO Data Using Invertible Neural Networks

3D Inversion of Large-Scale Frequency-Domain Airborne Electromagnetic Data Using Unstructured Local Mesh

An ATEM 1D inversion based on K-Means clustering and MLP deep learning

Two-dimensional magnetotelluric inversion method based on deep learning

3D Time-Domain Airborne Electromagnetic Inversion Based on Secondary Field Finite-Volume Method

Deep Learning Inversion of Electrical Resistivity Data

A fast 3-D inversion for airborne EM data using pre-conditioned stochastic gradient descent

Magnetic Anomaly Inversion Based on Convolutional Neural Network

Footprint-driven stochastic approximation inversion for time-domain airborne electromagnetic data

Efficient Bayesian inference using physics-informed invertible neural networks for inverse problems

Inversion of Gravity Anomaly Based on AlexNet Deep Neural Network

The Bayesian Inversion Method With a Surrogate Modeling Based on Neural Network for GATEM Data

Three-Dimensional Separate and Joint Inversions of Multi-Component Frequency-Domain Airborne Electromagnetic Data: Synthetic Model Studies