We propose the deep Gauss-Newton (DGN) algorithm. The DGN allows one to take into account the knowledge of the forward model in a deep neural network by unrolling a Gauss-Newton optimization method. No regularization or step size needs to be chosen; they are learned through convolutional neural networks. The proposed algorithm does not require an initial reconstruction and is able to retrieve simultaneously the phase and absorption from a single-distance diffraction pattern. The DGN method was applied to both simulated and experimental data and permitted large improvements of the reconstruction error and of the resolution compared with a state-of-the-art iterative method and another neural-network-based reconstruction algorithm.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1364/OL.484862 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Gauss Newton Method for Solving Variational Problems of PDEs with Neural Network Discretizaitons.
J Sci Comput
July 2024
School of Mathematical Sciences, Peking University, Beijing, China.
The numerical solution of differential equations using machine learning-based approaches has gained significant popularity. Neural network-based discretization has emerged as a powerful tool for solving differential equations by parameterizing a set of functions. Various approaches, such as the deep Ritz method and physics-informed neural networks, have been developed for numerical solutions.
View Article and Find Full Text PDFBeyond the basics: A deep dive into parameter estimation for advanced PBPK and QSP models.
Drug Metab Pharmacokinet
June 2024
Systems Pharmacology, Non-Clinical Biomedical Science, Applied Research & Operations, Astellas Pharma Inc., Ibaraki, Japan. Electronic address:
Physiologically-based pharmacokinetic (PBPK) models and quantitative systems pharmacology (QSP) models have contributed to drug development strategies. The parameters of these models are commonly estimated by capturing observed values using the nonlinear least-squares method. Software packages for PBPK and QSP modeling provide a range of parameter estimation algorithms.
View Article and Find Full Text PDFSVR-Net: A Sparse Voxelized Recurrent Network for Robust Monocular SLAM with Direct TSDF Mapping.
Sensors (Basel)
April 2023
School of Electronics and Information Engineering, Beihang University, Beijing 100191, China.
Simultaneous localization and mapping (SLAM) plays a fundamental role in downstream tasks including navigation and planning. However, monocular visual SLAM faces challenges in robust pose estimation and map construction. This study proposes a monocular SLAM system based on a sparse voxelized recurrent network, SVR-Net.
View Article and Find Full Text PDFImplicit Solutions of the Electrical Impedance Tomography Inverse Problem in the Continuous Domain with Deep Neural Networks.
Entropy (Basel)
March 2023
Department of Mathematics and Statistics, University of North Carolina at Charlotte, Charlotte, NC 28223, USA.
Electrical impedance tomography (EIT) is a non-invasive imaging modality used for estimating the conductivity of an object Ω from boundary electrode measurements. In recent years, researchers achieved substantial progress in analytical and numerical methods for the EIT inverse problem. Despite the success, numerical instability is still a major hurdle due to many factors, including the discretization error of the problem.
View Article and Find Full Text PDFWe propose the deep Gauss-Newton (DGN) algorithm. The DGN allows one to take into account the knowledge of the forward model in a deep neural network by unrolling a Gauss-Newton optimization method. No regularization or step size needs to be chosen; they are learned through convolutional neural networks.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!