查询词典 neural network

Through above study, the theoretical analysis of the inherent fault tolerance of feed-forward neural networks has been extended from hard-limit activation functions to differentiable activation functions; 2. By using above method, fault tolerance of discrete Hopfield feedback neural networks have been also analyzed. It means that the above method can be applied not only for feed-forward neural networks but also for feedback neural networks; 3. For the feed-forward neural network with the Sigmoid activation function, Chebyshev Inequality method was presented.

这项研究将以前仅限于对硬限幅作用函数前向神经网络固有容错性能的理论分析推广到具有可微作用函数的前向神经网络； 2 针对离散型Hopfield反馈神经网络，利用上面提出的方法，对其同样进行了容错性分析，得出许多有用的结论和计算公式，说明上面提出的方法不仅适用于前向神经网络，同时也适用于反馈神经网络； 3 针对具有Sigmoid作用函数的前向神经网络提出了一种切比雪夫不等式法。

For the data set with noises, a regularization intropolation method is proposed according to regularization theory. The relation between the regularization intropolation method and radial basis function method is analysed and structure of regularization neural networks is proposed. RBF neural network is introduced by mortifying the regularization neural networks. Finally the approximation capacity of RBF neural networks is analysed. 4. A method of selecting the centers of hidden layer neurons of RBF neural networks is proposed.

首先从精确内插问题开始对RBF神经网络进行讨论，然后根据正则化理论提出了在数据集带有噪声的情况下的内插方法，并分析了这种内插方法和径向基函数方法之间的密切联系以及其对应的正则化神经网络结构，其次对正则化神经网络进行了修改，得到正则化神经网络的简化形式—RBF神经网络，最后分析了RBF神经网络的逼近性能。

The whole PPT-reports of the Forum have put into "Complex Network Science Forum Circle" on Science Network in China (Please see: hettp//www.sciencene.cn/g/ComplexNets ) and also appear on the CCAST-WL Webside (Please see: www.ccast.ac.cn/workshop ). Main issues of the Forum are associated main subject of Network Science with application, including: review and outlook of network science ,spreading model of the H1N1 influenza virus , dynamics of neural network , brain functional networks,dynamics of topological structure, military network science with application, space structure of complex networks, community structure,,stationary e fficiency of coevolutionary networks , significance of community structure,,relationshipe between structure and dynamics of complex network, generalized synchronization and control, consensus of multi-agent systems, evolutionary game and friendship network, s preading of epidemic based on human and animal mobility pattern, complex network engineerings, and so on.

本届论坛文集共收编了32篇 ppt 报告，已经公布在中国科学网"复杂网络论坛圈"上（请看： hettp//www.sciencene.cn/g/ComplexNets 并且也发表在中国高等科学技术中心网站上（ the CCAST-WL Webside ）(请看： www.ccast.ac.cn/workshop )，这些报告内容丰富，各具特色，颇有创新，基本涵盖了近年来我国网络科学领域若干重要课题的研究进展和丰硕成果，课题研究颇有深度，不仅具有现实意义的重要课题（如甲型流感全球大爆发之经验统计及其传播模型），而且有前沿课题：生物神经网络系统的动力学与大脑网络，还有一系列重要课题：网络拓扑结构的动力学识别与估计、军事网络科学、网络科学在工程应用、多个体一致性、网络结构与动力学之间关系、同步-协同-中性稳定性和流型的稳定性、网络系统的协调动力学与控制、复杂网络的空间结构、连续系统和离散系统中的广义同步、疾病免疫算法、网络节点匹配、电力网、供应链网、化工网、通讯网与保密通信，等等。

And we used them into the inspection of these polluted gases. We used them because the constringent speed of wavelet neural network is very fast, they are not sensitive to the inputs of the network and they have the characteristic that they also can effectively approach the functions or signals. Classical neural networks mostly train the network with back propagation algorithm.

在目前常用的一维小波神经网络的基础上，我们研究了用来处理多维数据信息的小波神经网络，并将其应用于室内混合气体检测，主要是基于小波神经网络学习收敛速度较快，对网络输入不是很敏感，以及小波神经网络可以有效的进行函数逼近或者信号逼近的特点。

According to the actual circumstance of performance evaluation for mechanical products, the general neural network is reformed. Therefore, we construct the fuzzy neural network. This paper expresses membership function in the network by trapezium formulae .The indicators are fuzzed using the membership function .

网络中采用梯形函数的四元组形式构造模糊隶属函数，并用隶属函数的端点值和中间值作为样本，用改进的反向传播算法训练网络，得到与待评价产品情况相符合的权重值。

Reliability evaluation of existing structures is shown from the angle of artificial neural network. The influencing factors and deformation characteristics of buildings above goafs are summarized based on the code of coal pillar design and pressed coal mining for buildings, water-bodies, railroads and main mine roadway. The model of neural network is build to solve the problems about complex influencing factors and long evaluating time of building evaluation. BP network model compared with traditional way has strong practicability and extension value.

本文从人工神经网络的角度着手，基于国家煤炭工业《建筑物、水体、铁路及主要井巷煤柱留设与压煤开采规程》，分析了采动区建筑结构的影响因素和变形特性，运用已有的采动区建筑物可靠性评估实例，建立了BP人工神经网络可靠性评估模型，对采动区在役结构可靠性进行了评估，较好地解决了采动区建筑物可靠性评估影响因素复杂和耗时长的问题。

Then the self-recurrent neural network structure, one kind of simplified structure of internal recurrent network, and the leaming algorithm, dynamic backpropogation algorithm of self-recurrent neural network, is proposed.

本文将神经网络作为传统的线性模型的非线性推广进行了分析，证实了多层前向神经网络与非线性自回归模型的等价意义，及带有反馈的循环神经网络与非线性自回归移动平均模型的等价意义。

The deviation compensation algorithm put forward here is analyzed in theory and the effect of this compensation algorithm is testified based on experimental results that the learning speed of this network can be sped up markedly and all attitude magnetic heading error can be greatly reduced. RBF neural network is quite effective and superior to BP neural network.

在分析算法原理的基础上进行了实验仿真，结果表明：采用RBF神经网络在明显提高网络收敛速度的基础上，大大减小了全姿态磁航向误差，校正效果优于BP神经网络。

Nowadays, distributed intrusion detection system research based on the technology of agent and neural network becomes hotspot, software agent technology is considered as a very tool to develop intrusion detection system, especially its characteristic of Autonomy, Social Ability and Adaptability, and neural network technology gains more and more focus in respect that its pre-estimation to unknown network activities.

目前，基于Agent和神经网络技术的分布式入侵检测系统的研究成为业界热点，软件Agent技术因其所具有的良好的自治性、社会性、适应性，被视为在入侵检测系统中极为有用的技术之一，而神经网络技术则因其对未知网络行为的预判性，在入侵检测系统领域得到越来越多的关注。

The three-layered feedforward neural network and its learning methods are studied in this work, where new neural network model and the network learning algorithms are presented.

本文主要研究三层前向神经网络及其学习问题，在此基础上提出新的网络模型和学习算法，目的是为了克服基于传统的BP算法的前向网络存在的网络结构和。。。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。