Here I give some references to the selection of the number of hidden-layer nodes and learning rate.
在这里我们对于隐含层节点数选择、学习速率选择等问题提出一些参考意见。
参考来源 - 基于网络的智能监控系统的分析与设计Taking electro-hydraulic servo rudder-machine as object simulation, appropriate network structure, initial value of weight and learning speed was selected, self-corrected the weight of network to adapt changing of object controlled, robustness control of speeding-change process was realized.
以电液伺服舵机为对象仿真,选择合适网络结构、权值初值和学习速率,自动修正网络权值以适应变化的被控对象,实现了快变过程的鲁棒控制。
参考来源 - 基于改进Elman网络的自适应预测函数控制The learning rate is an important parameter for the learning process of a neural network(NN)which influents the stability and quickness of the NN.
学习速率是控制神经网络学习过程的一个重要参数,影响神经网络的稳定性和快速性。
参考来源 - 期刊学术社区·2,447,543篇论文数据,部分数据来源于NoteExpress
最常用的评估学习速率方法是绘制预测质量-项目个数的散点图。
The most common method to evaluate the learning rate is to plot a prediction quality versus number of items.
结果表明,本书中的方法比直观方法的学习速率快了1000倍。
The result indicates that it is about 1000 times faster than the intuitive method.
结果表明,网络的增益、学习速率和动量是影响网络收敛和稳定性的关键参数。
Restults show that the gain, learning rate and momentum are critical for network convergence and stability.
应用推荐