Deep unrolling networks with recurrent momentum acceleration for nonlinear inverse problems
From MaRDI portal
Publication:6557671
DOI10.1088/1361-6420/ad35e3MaRDI QIDQ6557671
Junqi Tang, Qing-Ping Zhou, Jinglai Li, Jiayu Qian
Publication date: 18 June 2024
Published in: Inverse Problems (Search for Journal in Brave)
recurrent neural networkinverse problemsmomentum accelerationdeep unrolling networkslearned primal-duallearned proximal gradient descent
Cites Work
- Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers
- A first-order primal-dual algorithm for convex problems with applications to imaging
- Learning nonlinear electrical impedance tomography
- Proximal Splitting Methods in Signal Processing
- Nonstationary inverse problems and state estimation
- Solving ill-posed inverse problems using iterative deep neural networks
- Construct Deep Neural Networks based on Direct Sampling Methods for Solving Electrical Impedance Tomography
- Computed tomography reconstruction using deep image prior and learned reconstruction methods
- Modern regularization methods for inverse problems
- High-Dimensional Mixture Models for Unsupervised Image Denoising (HDMI)
- Enhancing electrical impedance tomography reconstruction using learned half-quadratic splitting networks with Anderson acceleration
- A comparative study of variational autoencoders, normalizing flows, and score-based diffusion models for electrical impedance tomography
This page was built for publication: Deep unrolling networks with recurrent momentum acceleration for nonlinear inverse problems
