查询词典 neural network

In accordance with discrete, non-symmetry and time-delay feedback neural networks, we provide some theoretical criteria to judge the stability for designing the networks. On continuous feedback networks, we built an algorithm to estimate the convergence region. On the HNN feedback networks, an effective optimization algorithm based on the annealing strategy and the chaotic neural network is presented, and some disciplinary conclusions are provided to guide the selection of parameters. Simulation results on benchmark problems demonstrate that the performances of our algorithm are distinctly superior to those of the classic HNN based on gradient descend and stochastic optimization methods based on probabilistic distribution.

针对离散、非对称和时延等三种类型的动态反馈网络，给出了可用于网络设计的稳定性判别准则；针对连续动态反馈网络，建立了其收敛域估计的一个算法；针对以HNN模型描述的动态反馈网络，基于退火策略和混沌神经网络提出了一种高效的优化算法，建立了可用于指导参数选择的一些规律性结论，对典型算例的仿真研究显示，改进算法在各类性能上相对基于梯度下降的传统HNN模型和基于概率分布的随机优化算法，具有明显的优越性。

Generalization of neural network is a very important topic for coordinate transformation in neural computation.

在神经计算中神经网络的泛化特性是一个非常重要的内容。

This paper used Elman neural network to establish the model of irrigation water use prediction, having four importing units and exporting units. The determination of concealed level neural single number used the cut-and-try method, ten units were used.

运用神经网络中Elman反馈型神经网络建立了灌溉用水量预测模型，模型输入层神经元数目为4，输出层神经元数目为1，隐含层神经元数目确定采用试验法，最终确定为10。

Thirdly, in order to improve the learning efficiency and stability of feedforward neural network, a fast learning algorithm for neural networks based on CGM-OC approach is presented. Compared with other learning methods such as BP method, Broyden Flecher Goldfarl Shanno method and Powell method in which the derivative is replaced by the difference quotient etc, simulation results show that the proposed method is an efficient and fast one.

第三，为了提高前馈神经网络权值的学习效率及稳定性，提出一种基于正交校正共轭梯度优化方法的神经网络学习算法，通过与其它学习算法（如：BP算法、变尺度法、用差商近似代替导数的Powell法等）的比较，经仿真试验表明，本算法是一种高效、快速的学习算法。

So it is difficult to distinguish them using traditional neural network classification. A face distinguishment approach based on multi-stage neural networks is proposed.

为此，本文提出一种基于多级分类神经网络级联的人脸判别方法，通过把基于不同颜色分量的多级神经网络级联起来，并利用不同颜色分量由于来源、定义、算法的不同而体现出来的不同特性，以不同的颜色分量图象作为各级神经网络的分类特征，有效提高了人脸判别的正确率。

Some SAR image eigenspaces are constructed for different aspects based on independent component analysis. An independent neural network is trained for each of eigenspace of different aspects. The trained neural networks are used to recognize SAR targets, and their results are combined through D-S evidence reasoning.

通过独立分量分析构造若干个方位角的SAR图像特征空间，为每一特征空间各训练一个模糊极小极大神经网络用于分类，然后利用D-S证据理论集成各个神经网络的输出。

This dissertation uses Grey SystemTheory and Neural Network technology and suggests several grey neural networkmodels based on foregoer's works.

本文在前人的基础上，围绕系统工程的研究方法，将灰色系统理论与神经网络相结合，提出了灰色-神经网络方法。

Dynamical behaviors such as periodic, quasi-periodic and chaotic waveforms are observed on oscilloscope. It is expected that the proposed chaotic sources and their circuit implementations are beneficial to the application of neural networks. 4 A template of Cellular Neural Network is derived aimed at solving second-order hyperbolic wave equations based on space discrete method, a novel alternating recursive algorithm implementation is given.

提出了采用复数神经网络进行FIR滤波器设计的方法，利用复数信号能够同时携带幅度和相位信息的特点，构造了相应的具有复数的状态、偏置输入、输出、网络联接权的复数神经网络，证明了算法的收敛性，给出了网络的电路实现图，这方面工作对于拓展神经网络算法的应用范围和寻求FIR滤波器的设计新方法是有益的。

This approach is a combination of signal analysis and mathematical model .By the denoising function of wavelet and the learning itself function of neural network , the input and output nonlinear dynamic characteristic of system is obtained . The output prediction error, generated from the real output and wavelet neural networks estimated output, is used as a residual error to execute logical judge. and this approach can improve the speed and accuracy rate of fault detection.

它是一种把信号分析和数学模型相结合的故障检测方法，通过小波对信号的去噪和神经网络的自学习功能，来获取系统输入输出的非线性动力学特性，进而实时计算出残差并进行逻辑判决，可提高故障检测的速度和准确率。

As CMAC neural network could not be used to approximate the multi-valued mapping of an inverse hysteresis directly, an inverse hysteretic operator was proposed to transform the multi-valued mapping into a one-to-one mapping which could enable neural networks to approximate the behavior of an inverse hysteresis.

由于CMAC神经网络不能够直接逼近滞回逆这种具有记忆性的多映射现象，通过引入一个滞回逆算子，将多映射的滞回逆转换成一一映射，然后运用CMAC神经网络控制器来逼近这个一一映射，从而建立一个基于CMAC神经网络的滞回逆模型。

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。