查询词典 neural network

By strengthening leading of the party and the theory supporting; making the specialized capable and effective work troop of the network ideological and political education; displaying the function of the network carrier and strengthening the construction of the position of the network; establishing and perfecting the mechanism of the network ideological and political education in network The exploration in the view of study and systemic analysis for actual effect of network ideological and political education for resoling cyber crime. On this thesis have certain significance in theory and reality for promoting the innovation of network ideological and political education and strengthening of actual effect of it.

说到底互联网己不只是一种媒介、媒体和传播工具，它已经完全构成了一个新的社会形式，在这些认识主体、认识对象、教育主体、教育对象、教育方式、教育途径、教育形式以及教育内容都发生了前所未有的变化情况下，就必然要把社会发展、人的发展和思想政治教育的发展，把思想政治教育各个方面、各个环节的发展联系起来、整合起来，研究网络环境下的思想政治教育创新发展的系统性、规律性，来应对网络犯罪的严峻形势。

The author has suggested for the first time batch formula theory building on glass network structure concept basis.The upshots of batch theory are that intrasystem oxides was divided into network-forming oxide,network-modifying oxide and network-finishing oxide according to their property,and batch was computed according to quantitative rateof network-forming oxide,network-modifying oxide and network-finishing oxide.

笔者首次提出了建立在玻璃网络结构概念基础上的热法制作硅钾肥的配方理论，即将系统内氧化物组分按照其在玻璃网络体中的作用，划分为网络形成体和网络改性或修饰体，根据它们之间的量比关系进行配料计算。

A method for carrying out terminal management based on EPON includes connecting unit management system on optical line terminal based on EPON and setting up network management channel between them and between optical line terminal and each network terminal, carrying out analysis on message by optical line terminal when network management protocol message is sent to optical line terminal by unit management system, retransmitting message to network terminal by set up operation-maintenance channel when network protocol message is aimed at network terminal and retransmitting message reported from network terminal to unit management system by optical line terminal.

一种对基于EPON进行终端管理的方法，在基于EPON的光线路终端上连接设备管理系统，在它们之间建立网络管理通道，并在光线路终端和各网络终端之间建立承载在操作管理维护协议层上的网络管理通道；设备管理系统向光线路终端发送网络管理协议报文时，光线路终端对该报文进行解析，如果是针对网络终端的网络管理协议报文，则将该报文通过所建立的操作维护通道转发给该网络终端；网络终端向设备管理系统上报的消息，由光线路终端转发给设备管理系统。

The process of establishing neural network is as follows: two-layer neuronic network configuration is selected, S-type active function is used, input parameter of network are selected for thickness, preheat temperature, heat input and sampling point, output parameter of network is selected for fracture toughness property CTOD, learning algorithm of network is used for back-propagation algorithm.

本文选择了2层神经元的神经网络结构，采用S型激活函数，将板厚、预热温度、线能量、取样位置作为神经网络的输入，将环焊接头的裂纹尖端张开位移CTOD值作为网络的输出，通过BP算法的网络学习，采用VC语言建立了X70管线钢环焊接头CTOD与环焊工艺之间的神经网络。

The emulations and experimental results indicate that the theoretical model combined the neural network and genetic algorithm not only can conquer the disadvantage, which network is prone to enmesh the partial minimal error, but also utilize the powerful artificial intelligence of neural network.; This model provides a new method and idea for the monitor systems management, meanwhile it also offer a new method for solving the problems of the undetermined information processing; This model has powerful studying function, high accuracy, and strong practicability, so it has promoted value.

仿真及实验结果表明：由于此数学模型是将神经网络和遗传算法相结合应用在井下的监控系统中，因此即克服了神经网络容易陷入局部最小误差的缺点，又利用了神经网络本身强大的人工智能；该模型不仅为煤矿监控系统管理提供了一种新方法和新思路，同时也为解决不确定性信息处理问题提供了一种新的尝试和手段；该模型学习功能强大，精度较高，实用性强，具有很好的推广价值。

It employs the complex-valued step function to build neural model, then constructs network structure and decides the number of neurons; it adopts complex-valued Hebbian rules and inner product way to calculate the net weight matrix; then it analyzes the convergence of the network by taking advantage of the definition of energy function. In the image pre-processing stage, 2-dimensional Fourier Transform and Euler's formula are used, respectively, to transfer a gray-scaled image into phase information needed by the complex-valued neural network when stores the traffic sign. In the image post-processing stage, phase inverse transformation is used to change the phase image to gray-scaled image.

采用复数阶跃函数构建了神经元模型，建立了复数Hopfield神经网络结构，并且确定了网络中神经元的个数；借助复数Hebbian学习规则和复数内积法给出了网络权值的确定方法；借用复数网络能量函数的定义，说明了复数识别网络的收敛性；在前期的图像数据处理部分，分别使用离散二维傅里叶变换和欧拉公式，给出了将灰度图像数据转化为复数网络所需要的相位信息的方法，实现了网络对路牌的记忆存储；在后期的图像还原部分利用相位逆变换的方法，实现了相位图到灰度图转化。

The effects of the initial values and the network approximation error using neural network model composed of feedforward neural net and delayed output feedback in the approximation of NARMAX system are studied, and a conclusion is drawn that under some weak conditions the output of the network model has little difference from that of the actual system.

研究了用带有延迟环节的前馈网组成的网络模型逼近NARMAX系统时网络逼近误差及运行初始误差对网络模型输出的影响，指出在一些弱的条件下网络模型输出与实际系统输出间相差不大。

Considering the shortcoming of overlong network training time of BP neural network in multidimensional data classification,a new multiple neural network classification model is presented.

0引言高维数据的分类是一个重要而且具有挑战性的问题，基于BP神经网络的分类方法是一种有效的方法，具有较强的分类能力和鲁棒性。

To sum up, main results of this study were listed as follows:(1) After analyzing the main factors affecting generalization function of neural network, and aiming at poor representativity of hydrologic time series samples, standardization method of transforming extreme was presented to deal with input sample of time series, which could make the input samples lie in the interpolation range of training samples even if they lie in the extension of training samples, so it improved generalization function of neural network and accelerates network convergence speed.

概括起来，本文取得的主要研究成果如下：(1)对影响神经网络泛化能力的主要原因进行了阐述，针对水文时间序列样本的代表性不好这一问题，提出了采用变最值的规范化方法处理时间序列输入样本，使神经网络的输入样本即使为训练样本的外延，规范化后也处于训练样本的内插范围，从而改善神经网络的泛化能力，并加快网络收敛速度。

Self-generating neural network is a self-organization neural network, whose network structures and parameters need not to be set by users, and its learning process needs no iteration.

自生成神经网络是一类自组织神经网络，它不需要用户指定网络结构和学习参数，而且不需要迭代学习，是一类特点突出的神经网络。

Looking to bounce back from defeat against Cesena last weekend, the Calabrian boss will be sticking to his tried and tested 4-3-3, with Lopez and Sedivic joined up front in Giampaolo's absence by either Nanni or Plasmati.

鉴于上周末对切塞纳的比赛中反败为胜，这支南方球队将会更坚持其一贯而有效的4-3-3阵型，洛佩斯和塞德维奇将取代受伤缺席的詹保罗，和纳尼或者帕斯马蒂中的一个搭档锋线。

Meanwhile, their exposure to the more freewheeling ways of the West broadened dramatically.

另外，西方较自由的生活方式对年轻人的影响增大。