Solution of electromagnetic inverse problem using combinational method of Hopfield neural network and genetic algorithm

Kishimoto, Maki; Sakasai, K.; Ara, K.

doi:10.1063/1.360946

Cited by 52 publications

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The network energy of the Hopfield network decreases with network operation and so the total variation error for the inverse problem can be parameterized to the Hopfield network energy. However, this suffers from the problem that an optimum solution can not be obtained unless a suitable initial state has been set [ 8 , 9 ]. These techniques all require training of the neural network in the forward direction and are dual port networks (input and output).…”

Section: Introductionmentioning

confidence: 99%

Deep neural-network based optimization for the design of a multi-element surface magnet for MRI applications

2022

View full text Add to dashboard Cite

We present a combination of a CNN-based encoder with an analytical forward map for solving inverse problems. We call it an encoder-analytic (EA) hybrid model. It does not require a dedicated training dataset and can train itself from the connected forward map in a direct learning fashion. A separate regularization term is not required either, since the forward map also acts as a regularizer. As it is not a generalization model it does not suffer from overfitting. We further show that the model can be customized to either finding a specific target solution or one that follows a given heuristic. As an example, we apply this approach to the design of a multi-element surface magnet for low-field magnetic resonance imaging (MRI). We further show that the EA model can outperform the benchmark genetic algorithm model currently used for magnet design in MRI, obtaining almost 10 times better results.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep neural-network based optimization for the design of a multi-element surface magnet for MRI applications

2022

View full text Add to dashboard Cite

show abstract

“…To evaluate the performance of the simulated annealing method on experimental data, we tested it using laboratory data derived from a two-dimensional impedance change have been shown to have faster convergence speed than that of the simulated annealing algorithms in many apphcations [99][100][101][102][103][104][105][106][107][108][109]. We tested the genetic algorithm based approach using both synthesized impedance change data and laboratory data.…”

Section: Test Results For Laboratory Datamentioning

confidence: 99%

“…Applications of genetic algoritlims include configuration of radial basis function networks [102], image reconstruction [105], design and analysis of control systems [106], solution of electromagnetic inverse problems [107], optimal design of digital filters [108], and image segmentation [109].…”

Section: Background Of Genetic Algorithms Basic Conceptmentioning

confidence: 99%

Signal processing and image restoration techniques for two-dimensional eddy current nondestructive evaluation

Wang¹

View full text Add to dashboard Cite

This manuscript has been reproduced from the microfihn master. UMI fihns the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter &ce, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quali^ of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overiaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographs included in the original manuscript have been reproduced xerographically in this copy. Ifigher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

show abstract

Electron g‐factor determinations in Penning traps

2013

View full text Add to dashboard Cite

The magnetic moment of the electron, expressed by the g-factor in units of the Bohr magneton, is a key quantity in the theory of quantum electrodynamics (QED). Experiments using single particles confined in Penning traps have provided very precise values of the g-factor for the free electron as well as the electron bound in hydrogen-like ions. In this paper the status of these experiments is reviewed. The results allow testing calculations of higher order Feynman diagrams. Comparison of experimental and theoretical results for free and bound particles show no discrepancy within the limits of error, thus representing to date the most sensitive test of QED. Moreover, the g-factor provides a unique access to fundamental constants, as e.g. the electron mass or the fine structure constant.

show abstract

Solution of electromagnetic inverse problem using combinational method of Hopfield neural network and genetic algorithm

Cited by 52 publications

References 4 publications

Deep neural-network based optimization for the design of a multi-element surface magnet for MRI applications

Deep neural-network based optimization for the design of a multi-element surface magnet for MRI applications

Signal processing and image restoration techniques for two-dimensional eddy current nondestructive evaluation

Electron g‐factor determinations in Penning traps

Contact Info

Product

Resources

About