网络函数

Optimization arithmetic of objective function is introduced to determine the optimum combination of these prediction methods.

采用了一阶指数平滑模型、神经网络模型进行聚合物驱产量的预测和比较分析；引入了目标函数的评优算法来确定前述的预测方法的最佳组合策略。

Therefore,in this thesis, by using of the Lyapunov function ,Lasalles in-variance principle and the analytical technique to construct return mappings[34-36],we describe some important properties of the dynamic behaviors of two class of discrete-time model of two-cell cellular neural networks ,which includes stability,asymptotic stability,the existence and attractability of periodic solutions.

因此，本学位论文利用Lyapunov函数和Lasalle不变集原理以及利用分析技巧构造回复映射等方法对两类二元离散细胞神经网络模型的动力学性态进行定性分析，讨论了这些模型的稳定性、渐近稳定性、周期解的存在吸引性，全文共分三章。

Based on Radial Basis Function neural network automatic text classification systems.

详细说明：基于径向基函数的神经网络文本自动分类系统。

The proposed model is applied to finding complex eigenpairs and zeros of cmplex polynomials. 5. We study Baldi and Hornik's conjecture on Oja PCA neural algorithm, and give an energy function for one-unit Oja rule.

研究了Baldi和Hornik关于Oja主分量神经网络算法的一个猜想，给出了一个单元的Oja算法的能量函数，从而部分解决了这一猜想。

The fuzzy base function vector is introduced to approximate the upper bound of the uncertain vector.

在该方法中，利用模糊基函数网络在线学习系统不确定向量的上界。

Based on a bilinear transformation in multidimensional spaces, a neural-network based nonlinear analog fault diagnostic system for actual circuits is proposed.

本文提出了一种基于电路双线性函数在多维空间的变换与神经网络相结合的非线性模拟电路故障诊断新方法。

To obtain current-mode biquadratic filter with high accuracy, a voltage-mode second order universal filter based on two VFA's was transformed into a current-mode biquadratic filter-based on two AOA's by means of the concept of adjoint network.

为获得高精度电流模式双二次滤波器，以双VFA电压模式二阶通用滤波器为原型，用伴随网络的概念设计出AOA电流模式双二次滤波函数的新电路。

Standard neural network model is the interconnection of a linear dynamic system and a bounded static nonlinear operator, which is depicted as a linear differential inclusion to be analyzed easily with linear matrix inequalities technique.

标准神经网络模型(SNNM-Standard Neural Network Model)由一个线性动力学系统和有界激励函数构成的静态非线性算子连接而成，被表示为线性微分包含的形式，便于利用线性矩阵不等式技术进行分析。

The system consists of two main parts. One is status recognition system of calcine band flame, which is composed of image capturing, preprocessing, segmentation, feature extraction and pattern recognition. The other is neural network control system using Gaussian potential function network.

系统主要由两部分组成，一部分是回转窑煅烧带火焰燃烧状态识别系统，包括火焰图象获取、预处理、分割、特征提取与识别；另一部分是基于高斯基函数神经网络的控制系统。

Aiming at the problems of over-fitting and generalization for neural networks, the locally most entropic colored Gaussian joint input-output probability density function estimate is studied, and a new method by means of Chebyshev inequality is proposed to self-revise respectively according to every cluster.

针对神经网络的过拟合和泛化能力差的问题，研究了样本数据的输入输出混合概率密度函数的局部最大熵密度估计，提出了运用Chebyshev不等式的样本参数按类分批自校正方法，以此估计拉伸样本集，得到新的随机扩充训练集。

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。