Comparison of four neural net learning methods for dynamic system identification

Qin S.-Z.; Su H.-T.

首页> 外文期刊>IEEE Transactions on Neural Networks >Comparison of four neural net learning methods for dynamic system identification

【24h】

Comparison of four neural net learning methods for dynamic system identification

机译：四种用于动态系统辨识的神经网络学习方法的比较

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相似文献
相关主题

摘要

Four types of neural net learning rules are discussed for dynamic system identification. It is shown that the feedforward network (FFN) pattern learning rule is a first-order approximation of the FFN-batch learning rule. As a result, pattern learning is valid for nonlinear activation networks provided the learning rate is small. For recurrent types of networks (RecNs), RecN-pattern learning is different from RecN-batch learning. However, the difference can be controlled by using small learning rates. While RecN-batch learning is strict in a mathematical sense, RecN-pattern learning is simple to implement and can be implemented in a real-time manner. Simulation results agree very well with the theorems derived. It is shown by simulation that for system identification problems, recurrent networks are less sensitive to noise.

机译：讨论了用于动态系统识别的四种类型的神经网络学习规则。结果表明，前馈网络（FFN）模式学习规则是FFN批量学习规则的一阶近似。结果，模式学习对于非线性激活网络是有效的，前提是学习率很小。对于网络的递归类型（RecN），RecN模式学习不同于RecN批量学习。但是，可以通过使用较小的学习率来控制差异。虽然RecN批次学习在数学意义上是严格的，但是RecN模式学习很容易实现，并且可以实时实现。仿真结果与导出的定理非常吻合。仿真表明，对于系统识别问题，循环网络对噪声的敏感性较低。

著录项

来源
《IEEE Transactions on Neural Networks》 |1992年第1期|P.122-130|共9页
作者
Qin S.-Z.; Su H.-T.;
展开▼
作者单位

展开▼
收录信息
原文格式 PDF
正文语种 eng
中图分类自动化技术、计算机技术;
关键词

相似文献

外文文献
中文文献
专利

1. Dynamic System Identification by Neural Network (A New Fast Learning Method Based on Error Back Propagation) [J] . Chinmoy PAL, Naoki KAYABA, Shin MORISHITA, JSME International Journal. Series C, Mechanical Systems, Machine Elements and Manufacturing . 1995,第4期

机译：基于神经网络的动态系统识别（基于误差反向传播的新型快速学习方法）
2. Dynamic System Identification by Neural Network (A New Fast Learning Method Based on Error Back Propagation) [J] . Chinmoy PAL, Naoki KAYABA, Shin MORISHITA, JSME International Journal. Series C, Mechanical Systems, Machine Elements and Manufacturing . 1995,第4期

机译：基于神经网络的动态系统识别（一种新的基于误差反向传播的快速学习方法）
3. Identification of piecewise linear dynamical systems using physically-interpretable neural-fuzzy networks: Methods and applications to origami structures [J] . Liu Zuolin, Fang Hongbin, Xu Jian Neural Networks: The Official Journal of the International Neural Network Society . 2019,第期

机译：使用物理解释的神经模糊网络识别分段线性动力系统：对折纸结构的方法和应用
4. NEW METHOD OF DYNAMIC SYSTEM IDENTIFICATIONAND NEURAL NETWORKS LEARNING [C] . Ping-Yuan Cui, You-An Zhang, Masayuki Suzuki IFAC world congress . 1999

机译：动态系统识别和神经网络学习的新方法
5. Deep Neural Network Learning for Dynamic System Identification in Real-Time Using an Embedded GPU [D] . Autobee, Nicholas. 2021

机译：深度神经网络学习动态系统识别实时使用嵌入式GPU
6. Deep-learning: investigating deep neural networks hyper-parameters and comparison of performance to shallow methods for modeling bioactivity data [O] . Alexios Koutsoukas, Keith J. Monaghan, Xiaoli Li, 2017

机译：深度学习：研究深度神经网络的超参数并将性能与浅层方法进行生物活性数据建模的比较
7. Comparison of Four Neural Network Learning Methods Based on Genetic Algorithm for Non-linear Dynamic Systems Identification [O] . Dr. Rafid Ahmed Khalil 2012

机译：基于非线性动态系统识别遗传算法的四个神经网络学习方法的比较

Comparison of four neural net learning methods for dynamic system identification

摘要

著录项

相似文献

相关主题

期刊订阅