误差函数

基于鲁棒误差函数的BP算法

BP Algorithm Based on Robust Error Function

用计算机绘制不同x的相对误差函数图。

The relative error function diagram of different x in ERDS has been drawn by a computer .

基于双曲正切误差函数常数模算法的DSP实现

Hardware Implementation for Algorithm of Constant Modulus Based on Error Function of Hyperbolic Tangent

为避免这一现象，设计了一种新的误差函数结构，得到改进的BP算法。

In order to overcome it , a new error function structure and an improved BP algorithm were presented .

本文对基于不同误差函数小波盲均衡算法及DSP仿真实现进行了深入研究。

Wavelet blind equalization algorithm based different error function and simulation on DSP is deep studied in this dissertation .

由于样品未作高温热退火处理，为此由Al组分误差函数模型推导的扩散长度要大大高于扩散系数公式。

The diffusion lengths of Al component calculated from error function were larger than that calculated from coefficient of diffusion formula .

所以在BP网络的误差函数中加入惩罚项是提高网络泛化能力的一个重要途径。

So adding a penalty term to the error function of BP network is an important approach to gain better generalization .

BP神经网络采用梯度下降法来减小误差函数，可以同时研究大量的地震特征参数。

BP neural network use gradient descent to decrease error function , and to study a large number of seismic parameters .

该方法通过对加权误差函数的调节，实现了神经网络的等波纹一维FIR线性相位滤波器设计。

The approach can implement equiripple filter design by adjusting the weighted error function at iterations .

本文首先研究带有多个输出单元的BP神经网络的梯度学习算法，证明了误差函数在梯度算法所生成的权向量序列上的单调递减性，以及该算法的弱收敛性和强收敛性。

In this paper , we study the convergence of gradient algorithm for a three-layer BP neural network with multiple output units .

仿真结果表明：将K-L信息距离引入神经网络学习误差函数后得到的改进BP算法具有更好的泛化能力，可以进一步提高报价系统的泛化性能。

The simulation proves that the generalization ability of the system trained with the modified algorithm is much better than that of other algorithms .

误差函数erf（x）近似计算及其在开采沉陷预计中的应用

Proximate Calculation of Error Function erf ( x ) and Its Application in Mining Subsidence Prediction

直肠及乙状结肠癌CT评价&用水作造影剂神经元的转移函数采用sigmoid函数，误差函数采用二乘误差函数。

The Evaluation of Primary Rectal and Sigmoid Carcinoma by CT with Water as Contrast Agent The transfer function is sigmoid function .

提出一种新的泛化BP算法，在原误差函数中增加一项新的能量函数来提高系统的鲁棒性。

A new general learning BP algorithm is presented . By adding to a new power function in old error function , the system has stronger robustness .

讨论了BP小波神经网络在训练过程中减小误差函数时最优方向的确定和自适应调整学习率的方法。

The determination of the best direction of step length vector and the adaptive adjustment of training rate for BP wavelet neural network ( WNN ) are discussed .

神经元的转移函数采用sigmoid函数，误差函数采用二乘误差函数。

The transfer function is sigmoid function .

计算了余误差函数分布和高斯分布扩散层的霍尔系数与扩散层表面浓度NS的关系曲线。

, Curves giving R as function of surface concentration N , have been calculated for two forms of diffusive distribution , namely : the complementary error function and the Gaussian distribution .

然后应用BP网络作为卧式储液罐液位控制系统的控制方法，采用平方误差函数能使网络误差达到给定误差范围内。

Then apply the BP network as a horizontal tank level control system , control method , using squared error function enables the network error to a given error range .

结合菌体浓度变化范围大这一特点，将传统BP网络的误差函数进行了改进，并利用最优停止法对网络进行训练，避免了过拟合现象。

Combined with the characteristic that biomass can vary in a wide range , the error function of the traditional BP network is improved . Meanwhile , optimal stopping rule is used to avoid over-fitting .

采用信号四阶和六阶统计量提取信号特征，使用新设计的误差函数训练RBF神经网络，使得识别的效率和正确度得到了明显的改善。

The forth-order and sixth-order cumulants of received signal are adopted for features extraction while RBF neural networks with a new designed training cost function being used for classifier .

在神经网络的训练中大多用如下均方误差函数（MS）作为网络的训练目标函数，直到现在，大多数快速算法都是基于MS基础上得到的。

The mean squared error function is used extensively in the training of backpropagation neural networks . Until now , most of the fast learning algorithms were derived based on the MS error function .

可解决BP算法中误差函数对权值的偏导数计算问题以及执行器输出的脉动变化和饱和问题。

It can be used to solve two problems , that is , the evaluation of the partial derivative of error function with respect to weight in BP algorithm , and the fluctuation and saturation of the actuator output .

在分数低阶α稳定分布噪声环境下，对EP信号潜伏期变化检测算法中的误差函数进行连续非线性变换，可以有效地抑制分数低阶α稳定分布的噪声效应。

For the latency change detection of evoked potentials , the effect of a - stable noise can be restrained effectively through implementing appropriate non-linear transform to the error function .

提出了利用贪心算法更新Gabor小波特征模板的方法，通过对误差函数最小化实现自适应修正目标特征点位置。

The deformable Gabor wavelet character template is presented by using the esurience algorithm . This algorithm can modify position of target characters self adaptively by minimizing the error function .

首先研究了前馈神经网络，提出了基于鲁棒误差函数的BP神经网络的算法，并且验证了其对非线性系统逼近的有效性。

Firstly , studied feed-forward neural network and put forward a new algorithm on BP network , called BP Algorithm based on Robust Error Function ( BPAREF ), and the algorithm is proved to be effective for approaching nonlinear system .

在Matlab环境下对分集合并中的差模环特性、最大比合并特性以及均匀采样二阶数字锁相环误差函数高通特性进行了仿真，给出了仿真结果。

The performance of the differential mode loop , the characteristic of MRC combining and the high-pass characteristic of the error transfer function of the uniform sampling second-order DPLL are simulated in the Matlab , the simulation results are given .

由于步长在迭代过程中自适应进行调整，使误差函数E在超曲面上的不同方向按照步长向极小点逼近，实现了对目标函数的优化。

Due to the adaptive adjustment of step length in the iteration process , error function E approaches the minimum point in different directions on the super curved surface in accordance with the step length to realize the optimization of goal function .

本文对改进BP算法进行了分析，充分考虑误差函数、活化函数以及学习率对算法性能的影响，结合预测模型的特点提出了适合于预测模型的改进融合BP算法。

The author analyzes the Improvement Back-Propagation Algorithm and combines with the prediction model and takes full account the performance of error function , the activation function and the learn rate , And brings about his original idea on improving Back-Propagation Algorithm .

讨论了扩散微分方程在几种条件下的积分解，采用辛普生法计算误差函数erf（z）的定积分值，就2个不同浓度无限源的扩散状况，进行了数值计算。

A series of integration solutions for differential diffusion equations in several conditions were discussed . The value of error function erf ( z ) was calculated by the SIMPSON method .

该方法采用频域的最小平方误差函数使所求的增益特性逼近所希望的增益特性，其计算公式非常简单，并可设计任意形状增益特性的FIR数字滤波器。

This method approximates the desired gain characteristics using minimum square error function . Because of its very simple formulation , This method can be used to design FIR digital filters with the various shape 's gain characteristics .