查询词典 neural network

BP neural networks have been widely used in mode recognizing, image processing, system control etc A lot of study has been done on the training algorithm of BP neural networks But there are still many short comings To complete the training of the weights of a BP neural network, the key problem is how to overcome the staying at the local minimum points as well as how to increase the training speed based on this condition This paper comes up w...

BP神经网络被广泛应用于分类模式识别、图像处理和系统控制等领域人们对BP网络算法进行了许多的研究，但尚有其不足之处为完成其权的训练，问题的关键在于如何避免陷入局部极小及在此前题下如何提高学习速度为此，就如何选取学习率η和动量矩α提出了改进方案，并应用于数字识别，得到了较为满意的结果

ABSTRACT :Wavelet neural network possesses more degree of freedom and adaptivity than multi-layer FP neural network.

小波神经元比多层前馈神经网络具有更多自由度和更好的适应性。

ABSTRACT :Wavelet neural network possesses more degree of freedom and better adaptivity than multi-layer FP neural network.

内容摘要：摘要：小波神经元网络比多层前馈神经网络具有更多自由度和更好的适应性。

Based on the principle of BP neural network,a neural network model for predicting the properties of SBR compound was established by using the additional levels of several ingredients as the input variables and the corresponding tensile strengthes,elongations at break,tear strengthes and t90 as the output variables.

沈阳化工学院材料科学与工程学院，辽宁沈阳110142）摘要：根据BP神经网络原理，以SBR中的几种配料用量为输入变量，对应的胶料拉伸强度、拉断伸长率、撕裂强度和t90为输出变量，建立神经网络分析预报模型。

A radial basis function neural network, as well as a back propagation neural network for comparison, is trained and tested using 23 series of centrifuge model test data, 2 fullscale test data, and prototype date of a practical project.

收集23组挡墙离心模型试验数据，2组足尺试验数据，1组实际工程的破坏数据，共26组样本作为训练及检验样本，建立了可用于加筋挡墙设计高度预测的径向基函数网络及误差反传网络模型。

Also the convergence of temporal difference method is discussed. The method is put forward that Neural Network is used to implement multiaction Adaptive Heuristic Critic and Q-learning first, and then the five methods that can speed up the reinforcement learning are presented in allusion to slow learning speed of reinforcement learning system, they are quantisation of input space, local approximating neural network used to realize the reinforcement learning system, synthesis reinforcement learning algorithm, action model and prior knowledge.

针对强化学习系统学习速度较慢的问题，提出了提高强化学习速度的5种方法，即输入空间的量化方法、采用局部逼近神经网络实现强化学习系统方法、采用强化学习综合学习算法和利用动作模型及先验知识等方法，通过仿真实验验证了所提出方法的有效性。

Based on the summarization of the stateoftheart of the shortterm traffic flow prediction methods, two short-term prediction models of traffic flow that include bp neural network and rbf neural network are discussed in this paper.

在总结交通流短期预测方法发展趋势的基础上，分别介绍了基于常规的bp神经网络和基于rbf神经网络的交通流量短期预测模型，并重点研究rbf网络模型的预测性能，确定了关键参数sc的最优值。

Firstly, population partition methodology is presented by combining characteristics of both the cooperative co-evolutionary genetic algorithm and the neural network. Secondly, a novel mixed encoding method is also put forward to perform synchronal optimization of structure and weights of the neural network. Proper crossover and mutation operators were designed based on the encoding method. Next, a novel structure optimization technique is given according to the relationship among encoding, representation and cooperative group. Representative individual selecting and fitness evaluating are given.

首先，结合合作式协同进化遗传算法本身特性和神经网络特点，给出种群分割方法；其次，为了实现结构和权值的同步优化，提出一种新的混合编码方法，并根据该混合编码方法设计新的交叉和变异算子；然后，根据编码结构、代表个体和合作团体之间的关系，提出一种新的结构优化方法；再次，给出进化过程所需代表个体选择、适应度构造方法等。

The paper adopts the RBF neural network to fit ship lines at the first time and puts forward the method determining the RBF center point sets suitable for fitting ship lines. It shows that the method is of much higher learning speed compared with BP neural network, and is both practical and feasible based on the examination of the fitting precision by using the power function and circular function.

采用RBF神经网络拟合船舶型线，提出了适合船舶型线拟合的RBF中心点集的选取方式，与BP网络相比大大提高了学习速度和精度，用幂函数和圆函数检验本方法拟合曲线的精度，充分证明本方法拟合船舶型线是实用可行的。

For controlling the angle of pitch and velocity of rotate shaft of helicopter,the control performance of fuzzy PID,neural network PID and immune PID is compared each other.The simulation results show that neural network PID controller has the highest accuracy.

针对直升机俯仰角度控制和旋转轴速度控制需求，对模糊PID控制、神经网络PID控制和免疫PID控制在不同控制规律下的系统控制效果进行了对比研究。

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。