Abstract:Physics-guided deep learning is an important prevalent research topic in scientific machine learning, which has tremendous potential in various complex applications including science and engineering. In these applications, data is expensive to acquire and high accuracy is required for making decisions. In this work, we introduce an efficient physics-guided deep learning framework for the variational modeling of nonlinear inverse problems, which is then applied to solve an electrical impedance tomography (EIT) inverse problem. The framework is achieved by unrolling the proposed Anderson accelerated Gauss-Newton (GNAA) algorithm into an end-to-end deep learning method. Firstly, we show the convergence of the GNAA algorithm in both cases: Anderson depth is equal to one and Anderson depth is greater than one. Then, we propose three types of strategies by combining the complementary strengths of GNAA and deep learning: GNAA of learned regularization (GNAA-LRNet), where the singular values of the regularization matrix are learned by a deep neural network; GNAA of learned proximity (GNAA-LPNet), where the regularization proximal operator is learned by using a deep neural network; GNAA of plug-and-play method (GNAA-PnPNet) where the regularization proximal operator is replaced by a pre-trained deep denoisers. Lastly, we present some numerical experiments to illustrate that the proposed approaches greatly improve the convergence rate and the quality of inverse solutions.

What problem does this paper attempt to address?

### Problems Addressed by the Paper The paper aims to address the issues of data scarcity and high precision requirements in nonlinear inverse problems, particularly in the application of Electrical Impedance Tomography (EIT). Specifically, the paper proposes an efficient physics-guided deep learning framework based on the Anderson Accelerated Gauss-Newton (GNAA) algorithm and deep learning to improve the convergence speed and solution quality of inverse problem solving. ### Background and Motivation In scientific machine learning, solving inverse problems often faces high data acquisition costs and the need for high-precision decisions. Particularly in Electrical Impedance Tomography (EIT), it is very difficult to obtain real data because the conductive distribution inside the human body cannot be directly measured. Therefore, improving the accuracy of image reconstruction with limited training data is crucial for effective COVID-19 detection and other similar imaging applications. ### Method Overview The paper proposes a physics-guided deep learning framework based on the Gauss-Newton algorithm and Anderson acceleration (GNAA). This framework unfolds the GNAA algorithm into an end-to-end deep learning method, combining the advantages of physical models and data-driven approaches. Specifically, the paper proposes three strategies: 1. **GNAA-LRNet**: Learning the singular values of the regularization matrix through deep neural networks. 2. **GNAA-LPNet**: Learning the regularization approximation operator through deep neural networks. 3. **GNAA-PnPNet**: Replacing the regularization approximation operator with a pre-trained deep denoiser. ### Main Contributions 1. **Convergence Analysis**: Proving the convergence of the GNAA algorithm under different Anderson depths. 2. **Performance Improvement**: Validating the effectiveness of the proposed three methods in improving convergence speed and solution quality through numerical experiments. 3. **Flexibility and Generalization Ability**: These methods can automatically learn hyperparameters and data-driven regularization terms, and have good generalization ability on small datasets. ### Application Scenarios This method is particularly suitable for scientific and engineering applications where data is scarce but high-precision solutions are required, such as the application of Electrical Impedance Tomography (EIT) in medical and industrial fields. ### Conclusion By combining the advantages of physical models and deep learning, the proposed method performs excellently in nonlinear inverse problems, significantly improving the accuracy and stability of image reconstruction with limited training data. This provides new ideas and tools for solving complex inverse problems in practical applications.

Anderson Accelerated Gauss-Newton-guided deep learning for nonlinear inverse problems with Application to Electrical Impedance Tomography

Enhancing Electrical Impedance Tomography Reconstruction Using Learned Half-Quadratic Splitting Networks with Anderson Acceleration

Bayesian Inference and Deep Learning for Inverse Problems

Physics Embedded Weight-Sharing Neural Network for Electrical Impedance Tomography

A Two-Stage Imaging Framework Combining CNN and Physics-Informed Neural Networks for Full-Inverse Tomography: A Case Study in Electrical Impedance Tomography (EIT)

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

Implicit Solutions of the Electrical Impedance Tomography Inverse Problem in the Continuous Domain with Deep Neural Networks

Solving Electromagnetic Inverse Problem Using Adaptive Gradient Descent Algorithm.

A Reliable Deep Learning Scheme for Nonlinear Reconstructions in Electrical Impedance Tomography.

DeepEIT: Deep Image Prior Enabled Electrical Impedance Tomography.

Unsupervised Coordinate-Based Neural Network for Electrical Impedance Tomography

A Learning-Based Method for Solving Ill-Posed Nonlinear Inverse Problems: A Simulation Study of Lung EIT

Dominant-Current Deep Learning Scheme for Electrical Impedance Tomography

A Physics-Constrained Deep Learning-Based Image Reconstruction for Electrical Capacitance Tomography

Learning Nonlinear Electrical Impedance Tomography

A More General Electromagnetic Inverse Scattering Method Based on Physics-Informed Neural Network

On a Dynamic Variant of the Iteratively Regularized Gauss-Newton Method with Sequential Data.

Deep unrolling networks with recurrent momentum acceleration for nonlinear inverse problems

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

A Novel Deep Neural Network Method for Electrical Impedance Tomography

Deep Regularized Compound Gaussian Network for Solving Linear Inverse Problems