神经网络

A mathematical model of dynamic neural networks is built, which is shown to be equivalent to an affine nonlinear control system.

引 言 神经网络系统是在现代神经生物学和神经心理学的研究基础上，模仿人的大脑神经元的结构特征和功能特征而建立起来的一种非线性系统。

Aimed at the dynamic characteristics of disparity gradient neural network, it is concluded that the network is globally asymptotically stable.

着重考察视差梯度神经网络的动态性质，用能量函数的方法分析网络的稳定性，得出网络是渐近稳定的结论，并通过实验得到验证。

Since the character of network capability can also be influenced by the processing function of hidden neurons, a Chebyshev orthogonal neural network, in which the Chebyshev function is chosen for the hidden unit processing function, is proposed and the optimization algorithm under norm‖x‖〓 is investigated.

由于网络隐层神经元处理函数对网络性能也有重要影响，故将Chebyshev函数选为隐单元的处理函数而提出Chebyshev正交神经网络，并讨论了一致范数‖x‖〓意义下的优化算法。

Stochastic variables included driven accelerations, driven time, torques, frictions and damps were considered basically. In the first, Monte Carlo method was applied to generate stochastic variables and dynamical responds of mechanism. Secondly, the application of Artificial Neural Network was motivated by the approximate concepts inherent in reliability analysis and time consuming repeated analyses required for MC.

将驱动加速度、驱动时间、摩擦和阻尼力等作为随机变量，应用蒙特卡罗方法，取得动态参数样本，再利用人工神经网络方法，根据抽取的样本对网络进行训练，统计网络输出得到动态应力分布，进而求出机构动态强度可靠度。

Boltzmannn computer network is Hinton and others in 1985 will be simulated annealing primer access to the neural network, the n...

Boltzmannn机网络是Hinton等人在1985年将模拟退火算法引入到神经网络中，提出的，简称BM网络。

By taking temperature and horizontal coordinates as network input, and the rib-elevation as network output, the neural network model of rib-elevation was established.

此外，以气温、坐标和横坐标为网络输入，拱肋标高为网络输出，建立了拱肋标高的人工神经网络模型。

The RBF, therefore, outperformed the BP in terms of the prediction accuracy. The result shows that the RBFN and BPN can be successfully applied to build the relation of the incidence of stripe rust and meteorological data.

可见人工神经网络用于小麦条锈病的预测是可行的，并且逼近能力、分类能力和学习速度等方面均优于BP网络的径向基函数网络的预测效果更佳。

Initialization is a crucial problem in the application of multiwavelets, and the selection of initial weight affects directly the approximation performance and convergence rate of a multiwavelet neural network.

初始化问题是多子波神经网络的一个基本问题，网络初始值的选取直接影响着网络的逼近性能和收敛速度。

Since human's knowledge of real neural system is very limited, the improvement and development of artificial neural network need more detail information from neurophysiology and neuroanatomy.

由于人类对真实神经系统只了解非常有限一部分，人工神经网络的完善与发展有待于神经生理学、神经解剖学的研究给出更加详细的信息和证据。

This neural-fuzzy system can model a MIMO nonlinear system, its conclusion part of its fuzzy rules is ARX model.

提出了一种新的神经-模糊系统，该神经-模糊系统面向多入多出系统，结构上借鉴神经网络，共有六层。

We have no common name for a mime of Sophron or Xenarchus and a Socratic Conversation; and we should still be without one even if the imitation in the two instances were in trimeters or elegiacs or some other kind of verse--though it is the way with people to tack on 'poet' to the name of a metre, and talk of elegiac-poets and epic-poets, thinking that they call them poets not by reason of the imitative nature of their work, but indiscriminately by reason of the metre they write in.

索夫农 、森那库斯和苏格拉底式的对话采用的模仿没有一个公共的名称；三音步诗、挽歌体或其他类型的诗的模仿也没有——人们把&诗人&这一名词和格律名称结合到一起，称之为挽歌体诗人或者史诗诗人，他们被称为诗人，似乎只是因为遵守格律写作，而非他们作品的模仿本质。

The relationship between communicative competence and grammar teaching should be that of the ends and the means.

交际能力和语法的关系应该是目标与途径的关系。