.png){kind=icon}
返回
Neuro-Impedance Function for Electromagnetic Modeling
.png){kind=icon}
PRE
AI
DOI：10.1109/MMM.2024.3486588.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Artificial neural networks (ANNs) can predict electromagnetic (EM) responses under different parameters quickly and accurately while avoiding time-consuming simulations. Knowledge-based neural network (KBNN) techniques incorporating previous knowledge improves model learning and speeds up development. When accessing such knowledge is difficult, the KBNN can employ neurofunction approaches. The existing neurofunction methods tend to deal with the issues of order changing and parameter discontinuity simply from a mathematical perspective. This article is the latest addition to the review of neurofunction methods, which can be summarized as neuro-impedance function (neuro-IF) EM modeling technology. Unlike previous methods, the new neuro-IF solves the problem essentially by changing the physical meaning of the function, which fundamentally avoids the issues of order changing and parameter discontinuity. This new idea further expands the modeling capability boundaries of neurofunction methods and provide a new inspiration for future research.
Keyword:
Training
Reviews
Optimization methods
Predictive models
Feature extraction
Mathematical models
Vectors
Parametric statistics
Propagation constant
Usability
期刊
IF:
2.6
论文数:
3.2K
被引数:
3.1K
机构
引用论文
分享收藏LncRNA SNHG17 promotes gastric cancer progression by epigenetically silencing of p15 and p57LncRNA SNHG17通过表观遗传沉默p15和p57促进胃癌进展
PREAI
分享收藏Modeling of Voltage-Controlled Oscillators Including I/O Behavior Using Augmented Neural Networks.png)
IEEE ACCESS
OAAI
IEEE ACCESSIF3.6
OAAI分享收藏Hip Abductors And Adductors Explosive Capacity Correlate With Step Reaction Time In Older Adults
OAAI
OAAI分享收藏分享收藏
查看更多内容