RNN

The results exhibit good convergence and high accuracy of the network and the predictive capability is superior to RNN .

结果表明，双隐层过程神经网络收敛速度快、精度高，优于递归神经网络的预测结果。

Differentiable manifolds methods in RNN .

递归神经网络微分流形方法。

A New Method for Training RNN via Hidden Representation Estimated by EM Algorithm

利用EM算法估计隐含观测量的回馈神经网络学习新方法

RNN theoretical basis with the introduction of some Recurrent Neural Networks models .

介绍一些递归神经网络模型。

Multi-step Network Delay Prediction Model Based on RNN

基于随机神经网络的多步网络时延预测模型

The recurrent neural network ( RNN ) model based on projective operator is studied .

研究了一种基于投影算子的神经网络模型。

A Non-linear Prediction Speech Coding Based on RNN

基于RNN的非线性预测语音编码

Introduction : the basic concept of RNN .

绪论：递归神经网络的基本概念。

The simulation results demonstrate high modeling capabilities of RNN with simplest training data format for both of the cases .

仿真结果表明回归神经网络对时序关联数据有很好的建模和预测能力，相比于前向网络，无需过程时序特点的先验知识，可以采用最简单的建模数据形式。

The results of simulation indicated that RNN outperformed BP network and general median filter in image fusion for filtering .

仿真试验表明，粗集神经网络在图像融合滤波方面的性能优于BP网络和一般的中值滤波器。

Data format for modeling , capabilities of various RNN structures and evolutionary operations for RNN training are also investigated .

针对两个标准问题，采用不同形式的建模数据，比较了前向网络和回归神经网络的建模及预测效果，进一步将进化算法用于不同结构回归神经网络的训练并比较了它们的建模能力。

The network was compared with recurrent neural network ( RNN ) by predicting exhaust gas temperature ( EGT ) .

分别利用递归神经网络和双隐层过程神经网络对发动机排气温度裕度进行仿真预测。

The Identification and the Model of the Hydraulic & Pneumatic Actuator Based on RNN Recurrent Neural Networks linear-acting hydraulic power motor

基于RNN的神经网络辨识液气联合工作液压作动器模型线性作用液压电动机

It appears special construction , excellent short-term memory , facile learning approach , stunning non-linearity that former RNN never done .

其独特的结构，良好的短期记忆能力，方便的学习方法，不俗的非线性特性是以前循环神经网络所不可比的。

Elman recurrent neural network ( Elman RNN ) was applied to study the simultaneous quantitative analysis of overlapping UV spectra .

应用Elman回归神经网络对同时定量分析紫外重叠光谱进行了研究。

The query efficiency of RNN and NN has a high improvement , and there is a good performance especially in high-dimensional space query .

与其他算法的查询效率相比有了一定的提高，特别是在高维空间查询时，体现了良好的性能。

A small discrepancy is due to the fact that the RNN only approximates the first-principles CSTR model .

一个小的差异是由于这样的事实，RNNCSTR采用的仅仅是近似模型。

In this paper a detailed principle and application of a 1 / n rate convolution decoder based on Recurrent Neural Network ( RNN ) are introduced .

详细介绍了一种基于递归神经网络（RNN）的1/n卷积码解码器的原理与实现。

Combined with RNN ( reciprocal nearest neighbor ) algorithm , on relationship network Block-modeling , constructed the relationship of database information basic flow analysis . 3 .

结合最近邻聚类算法，对关系网络进行块模型分析，构建出了对数据库信息关系分析的基本流程。

This index structure has some advantages , greatly reducing the storage space for the quantitative and compression raw data , and the index is more suitable for the RNN than before .

这种索引结构的优点是对原始数据的量化压缩能够大大减少存储空间，引入的SR树的索引结构也更加适合反向最近邻查询的要求。

We demonstrated the application of modified RNN with self-feedback gains to rectify the simulated measurements obtained from dynamic continuous stirred tank reactor ( CSTR ) .

连续搅拌槽反应器实例的应用结果表明，反馈网络能快速、有效地校正动态过程测量数据。

By introducing the hidden representation or hidden variables into RNN , training the complicated RNN is decomposed into training a set of single neurons and a linear output layer .

利用隐含观测量，将复杂RNN的训练分解为线性输出层和多个单隐元的参数估计。

It has the characteristics of traditional RNN and the capability of function approximation , and may offer a kind of visual means for studying algebra symbol calculation and approximate algebra symbol calculation .

该模型不但具有回归神经网络的特点，而且具有Hensel构造提升的思想，给人们研究代数符号计算与近似代数符号计算提供一种可视化手段。

Furthermore , evolving algorithms , which can escape from local minima and get more suitable RNN models for the proposed problems , are more feasible for training various RNN structures .

而进化算法相比于常规的梯度下降算法，用于训练不同的回归网络结构通用性好，且训练过程不受局部极小问题的困扰，适当规模的训练过程可以获得性能良好的神经网络模型。

A new approach of short-term stock prediction using PSRT ( Phase Space Reconstruction Theory ) combined with RNN ( Recurrent Neural Network ) was presented according to the complex nonlinear character of stock time series .

根据股票指数时间序列复杂的非线性特性，提出以相空间重构理论与递归神经网络相结合的股票短期预测新方法。

For the problem of parameter variation and nonlinear dynamic friction compensation , a recurrent-neural-network ( RNN ) - based adaptive-backstepping control ( RNABC ) for servo system was proposed .

针对伺服系统的系统参数摄动和非线性动态摩擦补偿问题，提出基于递归神经网络（RNN）的自适应反步控制（RNABC）系统设计方法。

Random neural network ( RNN ) is a special kind of artificial neural network , which is developed recently and has its own peculiarities on the structure , the learning algorithm , the state-updating rule and the applications .

随机神经网络（RNN）在人工神经网络中是一类比较独特、出现较晚的神经网络，它的网络结构、学习算法、状态更新规则以及应用等方面都因此具有自身的特点。

In this study , we researched a well-known neural network modeling & MP modeling and suggest a new multidimensional MP modeling . As a biological neural mathematical model , RNN has particular advantages of associative memory , image processing and combinatorial optimization .

对著名的神经元数学模型MP模型进行了研究，提出了一种多维MP模型.作为仿生神经元数学模型，随机神经网络在联想记忆、图像处理、组合优化问题上都显示出较强的优势。

In light of the combination of research production on RNN and the scientific research project of the eighth oil recovery factory in Daqing oil fields , the sedimentary facies recognition system is developed based on RNN , which realizes auto recognition of sedimentary facies .

根据对粗神经网络的研究成果，并结合大庆油田采油八厂科研项目，研制开发出《基于粗神经网络的沉积微相识别系统》，实现了沉积微相的自动识别。

It was concluded by comparing three different preprocessing methods that the power spectra of alpha rhythm in wavelet transform domain could be used to predict epileptic seizures . The extraction of its envelope and nonlinear transform to the envelope provided further improvement of the performance of RNN effectively .

通过比较三种不同的预处理方法，发现在小波变换域利用脑电信号α节律的能量谱可以实现发作预报，而进一步提取包络并作非线性变换可以有效地提高RNN的预报性能。