# Physics-guided Loss Functions Improve Deep Learning Performance in Inverse Scattering

arXiv: https://arxiv.org/abs/2111.09109 · HTML: https://mlsanity.com/paper/2111.09109/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Builds on 3 of its 32 references that are in the map.
- First posted on arXiv 13 Nov 2021.

## Builds on (most cited first, 3 of 3)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Deep Convolutional Neural Network for Inverse Problems in Imaging](https://mlsanity.com/paper/1611.03679/) (2016)
- [DeepNIS: Deep Neural Network for Nonlinear Electromagnetic Inverse Scattering](https://mlsanity.com/paper/1810.03990/) (2018)

## Built on by (most cited first, 14 of 14)

- [Imaging Interiors: An Implicit Solution to Electromagnetic Inverse Scattering Problems](https://mlsanity.com/paper/2407.09352/) (2024)
- [Quality-factor inspired deep neural network solver for solving inverse scattering problems](https://mlsanity.com/paper/2504.20504/) (2025)
- [Physics-Driven Neural Network for Solving Electromagnetic Inverse Scattering Problems](https://mlsanity.com/paper/2507.16321/) (2025)
- [Multi-frequency Neural Born Iterative Method for Solving 2-D Inverse Scattering Problems](https://mlsanity.com/paper/2409.01315/) (2024)
- [Improved Physics-Driven Neural Network to Solve Inverse Scattering Problems](https://mlsanity.com/paper/2512.09333/) (2025)
- [Fast Physics-Driven Untrained Network for Highly Nonlinear Inverse Scattering Problems](https://mlsanity.com/paper/2602.13805/) (2026)
- [Aeroengine performance prediction using a physical-embedded data-driven method](https://mlsanity.com/paper/2407.00501/) (2024)
- [RINN: One Sample Radio Frequency Imaging based on Physics Informed Neural Network](https://mlsanity.com/paper/2504.15311/) (2025)
- [Enhancing Diffusion-Weighted Images (DWI) for Diffusion MRI: Is it Enough without Non-Diffusion-Weighted B=0 Reference?](https://mlsanity.com/paper/2505.12978/) (2025)
- [Electromagnetic Inverse Scattering from a Single Transmitter](https://mlsanity.com/paper/2506.21349/) (2025)
- [Hardware Software Optimizations for Fast Model Recovery on Reconfigurable Architectures](https://mlsanity.com/paper/2512.06113/) (2025)
- [Contrast-Source-Based Physics-Driven Neural Network for Inverse Scattering Problems](https://mlsanity.com/paper/2601.19243/) (2026)
- [Coordinate-Residual Physics-Driven Neural Network for Inverse Scattering Imaging](https://mlsanity.com/paper/2608.09382/) (2026)
- [Structured Multi-Level-Set Inversion for 3-D Electromagnetic Inverse Scattering](https://mlsanity.com/paper/2609.08594/) (2026)

## Abstract (arXiv)

Solving electromagnetic inverse scattering problems (ISPs) is challenging due to the intrinsic nonlinearity, ill-posedness, and expensive computational cost. Recently, deep neural network (DNN) techniques have been successfully applied on ISPs and shown potential of superior imaging over conventional methods. In this paper, we analyse the analogy between DNN solvers and traditional iterative algorithms and discuss how important physical phenomena cannot be effectively incorporated in the training process. We show the importance of including near-field priors in the learning process of DNNs. To this end, we propose new designs of loss functions which incorporate multiple-scattering based near-field quantities (such as scattered fields or induced currents within domain of interest). Effects of physics-guided loss functions are studied using a variety of numerical experiments. Pros and cons of the investigated ISP solvers with different loss functions are summarized.