MétaCan
Menu
← Back to cohort
Record W4386287401 · doi:10.1002/9781119853923.ch4

Artificial Neural Networks for Parametric Electromagnetic Modeling and Optimization

2023· other· en· W4386287401 on OpenAlexaff
Feng Feng, Weicong Na, Jing Jin, Qi‐Jun Zhang

Bibliographic record

Venuenot available
Typeother
Languageen
FieldEngineering
TopicMicrowave Engineering and Waveguides
Canadian institutionsCarleton University
Fundersnot available
KeywordsArtificial neural networkComputer scienceParametric statisticsParametric modelArtificial intelligenceExtrapolationGeneralizationMachine learningTransfer functionEngineeringMathematics

Abstract

fetched live from OpenAlex

This chapter introduces both the fundamentals and advanced formulations of artificial neural network (ANN) techniques for parametric electromagnetic (EM) modeling and optimization. ANN is an information handling system whose design was enlightened by the investigation into the human brain's capacity to learn from observations and to summarize through abstraction. ANN is an acknowledged tool for parametric EM modeling and optimization, that is, using geometric parameters as variables to represent the EM behavior. Direct methods for EM design optimization are generally computationally expensive and require repeated EM evaluations due to constantly changing geometry. ANN has become an effective method for EM parametric modeling by learning the relationship between EM conducts and geometric parameters. The ANN after training can quickly solve the EM behavior of microwave devices when the geometric parameters change repeatedly. If a neural network has multiple hidden layers, it is defined as a deep neural network. In modeling highly complicated sophisticated relationships, such as modeling with high-dimensional filters with numerous input variables, deep neural networks can do better than shallow neural networks (neural networks with only a few hidden layers). Exploiting the availability of prior knowledge for parametric EM modeling, knowledge-based neural networks (KBNNs) have been exploited. Compared with traditional ANN, KBNN can achieve identical modeling precision with fewer training data and offer preferable extrapolation, thus accelerating model development and improving the generalization ability of parametric EM modeling and optimization. A progressive knowledge-based modeling method, combining neural networks and transfer functions (neuro-transfer functions or neuro- transfer function [ TF s]), has been exploited for parametric modeling of EM responses. Since an appropriate equivalent circuit model/experience model may not be exploitable in certain cases, the neuro-TF approach is capable of using the transfer function as the prior knowledge. The ANN-based parametric EM models can be further utilized as surrogate models for EM optimizations. The exploration of ANN techniques for parametric EM modeling and optimization is a hot topic and continues to be an open and strategic direction.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.003
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.006
Threshold uncertainty score0.019

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.003
Meta-epidemiology (narrow)0.0020.001
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0010.001
Science and technology studies0.0000.001
Scholarly communication0.0020.001
Open science0.0010.001
Research integrity0.0020.003
Insufficient payload (model declined to judge)0.0060.002

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.013
GPT teacher head0.208
Teacher spread0.195 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2023
Admission routes1
Has abstractyes

Explore more

Same topicMicrowave Engineering and Waveguides→French-language works237,207→