Question

求电子信息工程专业英语翻译大神帮我翻译这段话

目前，关于神经网络的定义尚不统一，按美国神经网络学家Hecht Nielsen 的观点，神经网络的定义是：“神经网络是由多个非常简单的处理单元彼此按某种方式相互连接而形成的计算机系统，该系统靠其状态对外部输入信息的动态响应来处理信息的。”综合神经网络的来源、特点和各种解释，它可简单表达为：神经网络是一种旨在模仿人脑结构及其功能的信息处理系统。

从神经网络的理论模型来看，它主要可分为两大类；即层状前馈神经网络和互联反馈神经网络，对于层状前馈神经网络，它的特点是通过适当的BP算法进行样本训练学习，神经网络的输入输出可以逼近任意输入输出对应的非线性映射，这种经过学习来实现映射其实就是一种自适应的判断控制功能。由于层状前馈神经网络具有“自学习”和“自训练”的功能，并可模仿人类大脑的智能，因而具有很强的分类、识别能力。因此，层状前馈神经网络被广泛地用于网络通讯中的信道均衡、全局性网络管理、信息流量预测以及其它自适应控制等方面。而互联反馈神经网络的特点是通过设计学习将联想记忆内容或最优化答案设置成系统能量函数的极小点，经神经网络的动力学平衡过程就可以实现自动快速处理优化问题，因而它可以广泛地用于网络通讯领域，包括信息包的调度、最优路由的选择、信息交换和控制。高速互联的、非线性的神经网络还具有混沌行为，它是一个非常复杂的NP问题，能产生无法预测的序列轨迹，可以设计成安全可靠的快速密码算法。正是由于神经网络的学习映射、联想优化功能和混沌行为等特点能够在理论上解决目前宽带网络通讯技术所面临的一些问题，因而在网络通讯中得到广泛的应用。

3 神经网络在网络通讯中的应用

以下根据神经网络的“自学习”功能、联想优化功能和混沌行为这三个功能特点从三个方面介绍神经网络在网络通讯中的应用实例：

Answer 1

At present, the definition of neural network are not unified, according toAmerican neural network expert Hecht Nielsen point of view, the definition of neural networks is: "neural networks is a computer system very simpleprocessing units to each other in some way connected to each other and form,the dynamic response of the system depends on external input informationstate to process information." The sources, characteristics and varioussynthetic interpretation of neural network, it can be simply expressed as:neural network is an information processing system to imitate the human brain structure and function.
From the theoretical model of neural network, it can be divided into two categories; namely, and interconnection layered feedforward neural networkfeedback neural network, for a layered feedforward neural network, which is characteristic of sample training by BP algorithm appropriate, non-linear mapping of input and output of neural network can approximate any input and output the corresponding, this after learning to achieve the mapping is a kind ofadaptive judgment control function. Due to the layered feedforward neuralnetwork with "learning" and "self training" function, intelligent and can imitate the human brain, so it has very strong recognition ability, classification.Therefore, the layered feedforward neural network is widely used in network communications, channel equalization of global network management,information flow and other adaptive control. Characteristics andinterconnected feedback neural network is that associative memory content oroptimal answer set to the minimum energy function through the design oflearning, the dynamic balance process neural network can realize automaticfast processing optimization problem, so it can be widely used in the field of network communication, including scheduling, packets, optimal routing the exchange of information and control. A neural network nonlinear high-speedInternet, also has the chaotic behavior, it is a very complex NP problem, can produce a sequence of path cannot be predicted, the fast algorithm can be designed into a safe and reliable. It is because of the characteristics oflearning, neural network associative mapping optimization function and chaotic behavior can solve some problems of broadband network communicationtechnology faces in theory, which have been widely used in the communication network.
Application of 3 neural network in network communication
The neural network in accordance with the "self-learning" function, Lenovooptimization function and chaotic behavior of these three features from three aspects introduced the neural network in the communication networkapplication:

Answer 2

At present, about the definition of neural network are not
unified, according to the neural network at Hecht, Nielsen's point of view, the
neural network is defined as: "the neural network is made up of a number of very
simple processing units to each other in some way connected to form a computer
system, the system depends on its status, the dynamic response of the external
input information to process information." The sources and characteristics of
integrated neural network and a variety of interpretation, it can be expressed
as: simple neural network is a designed to mimic the human brain structure and
function of information processing system.From the perspective of the
theory model of neural network, it can be divided into two main categories;
Namely layered feedforward neural network, neural network and Internet feedback
for layered feedforward neural network, its characteristic is through
appropriate BP algorithm for sample training and learning of the neural network
input and output can approximate arbitrary input and output the corresponding
nonlinear mapping, that after learning to implementing a map is a kind of
adaptive control function. Due to the layered feedforward neural network has
"self learning" and "training" function, and can imitate the intelligence of the
human brain, thus has strong classification and recognition. Therefore, layered
feedforward neural network is widely used in network communication channel
equalization, the global network management, information flow prediction as well
as other adaptive control, etc. And the characteristics of the feedback neural
network is through design to study the associative memory content or
optimization answer set to minimum point of the system energy function, the
dynamic equilibrium process of neural network can realize the automatic rapid
processing optimization problems, so it can be widely used for the field of
network communication, including the selection of packet scheduling and optimal
routing, exchange of information and control. High speed Internet, nonlinear
neural network also has the chaotic behavior, it is a very complex NP problem,
can produce unpredictable sequence trajectory, fast password algorithm can be
designed to be safe and reliable. It is because of the neural network learning
mapping, lenovo optimization function and the characteristics of chaotic
behavior can in theory to resolve the broadband network communication technology
is facing some problems, which has been widely used in network
communication.3 neural network application in network
communicationThe following according to the "self learning" function of
neural network, lenovo optimization function and chaotic behavior of the three
functional features from three aspects: introduces the application of neural
network in network communication instance
求采纳