文章提出一种基于多分辨率学习的正交基小波神经网络结构的设计方法,网络权值的学习采用阻尼递推最小二乘算法。
A designing method of wavelet neural network structure based on multiresolution learning is put forward, and the studies of network weights adopt damped least squares.
R BF网络的设计问题就是关于网络隐节点数和隐层节点RBF函数中心、宽度和隐层到输出层的权值的性能指标的最小化问题。
The RBF network configuration is formulated as a minimization problem with respect to the number of hidden layer nodes, the center locations and the connection weights.
提出一种基于正交基函数的小波神经网络设计方法,采用多分辨率学习确定隐含层结构,并用收敛较快的阻尼最小二乘法训练权值。
In this approach the network structure is determined by multiresolution learning, and the weights are trained by damped least squares which has fast convergent rate.
设计要求:利用Prims算法求网的最小生成树;以文本形式输出生成树中各条边以及它们的权值。
Design requirements: the use of algorithms for network Prims minimum spanning tree to the text of the various forms of output spanning tree edges and their weights.
再用振动信号记录系统记录此时实验激励的振动加速度的均方根值。将记录的数据利用最小二乘和优化设计的方法拟合成振动频率计权曲线。
The data were fitted to the same form of the frequency weighting Wk of ISO 2631-1 with the least square method and optimization of parameters.
再用振动信号记录系统记录此时实验激励的振动加速度的均方根值。将记录的数据利用最小二乘和优化设计的方法拟合成振动频率计权曲线。
The data were fitted to the same form of the frequency weighting Wk of ISO 2631-1 with the least square method and optimization of parameters.
应用推荐