Results 21 to 30 of about 16,083 (234)
High-Resolution Diffusion-Weighted Imaging With Self-Gated Self-Supervised Unrolled Reconstruction. [PDF]
Magn Reson MedABSTRACT Purpose High‐resolution diffusion‐weighted imaging (DWI) is clinically demanding. The purpose of this work is to develop an efficient self‐supervised algorithm unrolling technique for submillimeter‐resolution DWI. Methods We developed submillimeter DWI acquisition utilizing multi‐band multi‐shot EPI with diffusion shift encoding.
Tan Z +4 more
europepmc +2 more sources
A Weberized Total Variance Regularization-based Image Multiplicative Noise Model
Image Analysis and Stereology, 2023 This paper considers Weber's law and proposes a new non-convex model for images contaminated by Gaussian noise and Rayleigh noise. The alternating direction method of multipliers (abbreviated as ADMM) is a recent popular method that can handle convex and
Xinyao Yu, Donghong Zhao
doaj +1 more source
Многоблочный метод ADMM с ускорением Нестерова
Міжнародний науково-технічний журнал "Проблеми керування та інформатики", 2021 Метод ADMM (alternating direction methods of multipliers) широко используется для решения многих оптимизационных задач с помощью параллельных вычислений.
Владислав Анатолійович Григоренко +2 more
doaj +1 more source
Convergence Analysis of Multiblock Inertial ADMM for Nonconvex Consensus Problem
Journal of Mathematics, 2023 The alternating direction method of multipliers (ADMM) is one of the most powerful and successful methods for solving various nonconvex consensus problem.
Yang Liu, Yazheng Dang
doaj +1 more source
Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems, 2019 To facilitate efficient embedded and hardware implementations of deep neural networks (DNNs), two important categories of DNN model compression techniques: weight pruning and weight quantization are investigated. The former leverages the redundancy in the number of weights, whereas the latter leverages the redundancy in bit representation of weights ...
Ao Ren +7 more
openaire +2 more sources
Horizontal Layered Scheduling ADMM Penalized Decoder Based on the Improved Penalty Function for LDPC
IEEE Access, 2022 For low-density parity-check (LDPC) codes, reducing the number of Euclidean projections, choosing a suitable scheduling strategy, and devising an improved penalty function are three effective ways to increase the alternating direction method of ...
Biao Wang, Zhongfei Wang
doaj +1 more source
Randomly assembled cyclic multi-block ADMM: a fast method for large-scale linearly constrained quadratic optimization [PDF]
, 2022 This work is motivated by a simple question: how to find a relatively good solution to a very large optimization problem in a limited amount of time. We consider the linearly constrained convex minimization model with an objective function that is the ...
Mihic, Kresimir
core +1 more source
The exact worst-case convergence rate of the alternating direction method of multipliers [PDF]
Mathematical programming, 2022 Recently, semidefinite programming performance estimation has been employed as a strong tool for the worst-case performance analysis of first order methods.
M. Zamani +2 more
semanticscholar +1 more source
An ADMM-based SQP method for separably smooth nonconvex optimization
Journal of Inequalities and Applications, 2020 This work is about a splitting approach for solving separably smooth nonconvex linearly constrained optimization problems. Based on the ideas from two classical methods, namely the sequential quadratic programming (SQP) and the alternating direction ...
Meixing Liu, Jinbao Jian
doaj +1 more source
Learning-based accelerated sparse signal recovery algorithms
ICT Express, 2021 In this paper, we propose an accelerated sparse recovery algorithm based on inexact alternating direction of multipliers. We formulate a sparse recovery problem with a concave regularizer and solve it with the relaxed and accelerated alternating method ...
Dohyun Kim, Daeyoung Park
doaj +1 more source