Location:
Search - parallel neural
Search list
Description: 对具有随机噪声的二阶系统的模型辨识，进行标幺化以后系统的参考模型差分方程为： y(k)=a1*y(k-1)+a2*y(k-2)+b*u(k-1)+s(k) 式中，a1=0.3366,a2=0.6634,b=0.68,s(k)为随机噪声。由于神经网络的输出最大为1，所以，被辨识的系统应先标幺化，这里标幺化系数为5。采用正向建模(并联辨识)结构，神经网络选用3-9-9-1型，即输入层i，隐层j包括2级，输出层k的节点个数分别为3、9、9、1个；由于神经网络的最大输出为1，因此在辨识前应对原系统参考模型标么化处理，辨识结束后再乘以标么化系数才是被辨识系统的辨识结果。-of random noise with the second-order system model, per-unit system after the reference model differential equation : y (k) = y* a1 (k-1) a2* y (k-2) b* u (k-1) s (k)- style, = 0.3366 a1, a2 = 0.6634, b = 0.68, s (k) as random noise. Because the neural network for a maximum output, therefore, the identification system should be per-unit, per-unit here coefficient of 5. Forward modeling (Parallel identification) structure, neural network-based selection 3-9-9-1, i input layer, hidden layer, including two j, k output layer to the number of nodes 3,9,9,1; The neural network the biggest losers up to one, in the original deal before Identification System Reference Model S Mody treatment, then multiplied by the end of Identification Standard Mody coefficient was recognition system is the ide
Platform: |
Size: 874496 |
Author: 孙荣超 |
Hits:
Description: 双并联人工神经网络,具有收敛速度快,泛化能力强等特点-double parallel artificial neural network, with the convergence speed and ability to generalize the characteristics of
Platform: |
Size: 269312 |
Author: 田新志 |
Hits:
Description: 6。《非数值并行算法——遗传算法》，刘勇等着 科学出版社 1995年第一版
本书系统地叙述了非数值并行算法之一的模拟退火算法的基本原理以及最新进展，同时为了便于读者解决实际问题，书中对具体算法的步骤作了详细介绍。本书共分七章，第一章介绍算法的思想、特点。发展过程和前景。第二章介绍算法的基本理论。第三章讨论算法解连续优化问题。第四章利用算法设计和优化神经网络。第五章介绍在组合优化中的应用。第六章介绍应用遗传程序设计解决程序设计自动化问题。第七章对遗传算法和其它适应性算法进行比较。
本书可供高校有关专业的师生、科研人员、工程技术人员阅读参考。-6. "Non- numerical parallel algorithms-- Genetic Algorithm" Liu Yong waiting 520-531 1995 version of the first book to systematically describe the non- numerical parallel algorithms one count of simulated annealing Law and the basic tenets of the latest progress, in order to help readers solve practical problems, book of the specific algorithm steps in detail. The book is divided into seven chapters, the first chapter describes the idea algorithm, characteristics. The development process and prospects. The second chapter describes the basic algorithm theory. The third chapter discusses Algorithm for continuous optimization problems. Chapter IV using the algorithm design and optimization of neural networks. The fifth chapter in combinatorial optimization applications. Chapter V
Platform: |
Size: 4088832 |
Author: 孙东 |
Hits:
Description: 8。《遗传算法及其应用》，陈国良等编着 人民邮电出版社 2001年第一版
本书系统全面地介绍了遗传算法的基本原理、设计方法及其并行实现，以及它在组合优化、机器学习、图像处理、过程控制、进化神经网络、模糊模式识别和人工生命等方面的应用。 本书可作为高等院校计算机、无线电电子学、自动控制、生物医学工程等有关专业高年级学生或研究生的教材和参考书，也可供从事人工智能、信息处理研究和应用的科技人员学习参考。-8. "Genetic algorithm and its application," Chen Guoliang edited and published with the People's Post and in 2001 the first book version of the full introduction of the genetic algorithm basic tenets , design methods and parallel implementation, and its portfolio optimization, machine learning, image processing, process control, evolutionary neural networks, Fuzzy Pattern Recognition and Artificial Life, and so on the application. The book can be used as institutions of higher learning computers, radio electronics, automatic control, biomedical engineering, and other professional grade students or graduate teaching and reference materials, are also available in artificial intelligence, information processing and application of information technology to study.
Platform: |
Size: 6904832 |
Author: 孙东 |
Hits:
Description: 经网络提出了一种基于蚁群聚类算法的径向基神经网络. 利用蚁群算法的并行寻优特征和挥发系数方法的自适应更改信息量的能力,并以球面聚类的方式确定了径向基神经网络中基函数的位置, 同时通过比较隐层神经元的相似性、合并相似性较为接近的2 个神经元来约简隐含层的神经元,以达到简化径向-via the Internet presents a clustering algorithm based on the ant colony RBFNN. Ant use the parallel algorithm optimization Volatile coefficient characteristics and the method of adaptive changes in the volume of information, spherical cluster and the method of RBF neural network-based functions of the position, By comparing the same time hidden layer neurons of similarity, Merger similarity is closer to the two neurons to Jane about the hidden layer neurons to achieve simplification Radial
Platform: |
Size: 31744 |
Author: 何泼 |
Hits:
Description: 提出了利用FPGA的现场可编程以及可并行处理的特性，对基于人工神经网络的图像处理结构进行自动生成的一种技术。作者：Andre B. Soares, Altamiro A. Susin，Leticia V. Guimaraes-Made use of field-programmable FPGA, as well as the characteristics of parallel processing, artificial neural network-based image processing to automatically generate the structure of a technology. Author: Andre B. Soares, Altamiro A. Susin, Leticia V. Guimaraes
Platform: |
Size: 1055744 |
Author: Rae |
Hits:
Description: 细胞神经网络(CNN)GUI源代码
细胞神经网络(CNN)是一种和人类神经网络非常相似的并行计算模型，各个邻接节点间有不同的通信。在本程序中A模型是反馈矩阵，B是控制矩阵。
-Cellular neural network (CNN) GUI source code for cellular neural network (CNN) is a human neural network is very similar to the parallel computing model, all adjacent nodes have different communication. In this process, the feedback matrix A model, B is the control matrix.
Platform: |
Size: 51200 |
Author: lcp |
Hits:
Description: 传算法的基本原理、设计方法及其并行实现，以及它在组合优化、机器学习、图像处理、过程控制、进化神经网络-Propagation algorithm of the basic principles, design methods and their parallel implementation, as well as in combinatorial optimization, machine learning, image processing, process control, evolutionary neural network
Platform: |
Size: 9216 |
Author: 鲁明 |
Hits:
Description: 介绍遗传算法基本原理和人工神经网络,并行遗传,遗传程序等,并用这些方法为商旅问题等求解-Genetic algorithm and Artificial neural network
Parallel GA
Genetic programming
Traveling salesman
Platform: |
Size: 135168 |
Author: 张庭 |
Hits:
Description: 人工神经网络(Aartificial Neural Network，下简称ANN)是模拟生物神经元的结构而提出的一种信息处理方法。早在1943年，已由心理学家Warren S.Mcculloch和数学家Walth H.Pitts提出神经元数学模型，后被冷落了一段时间，80年代又迅猛兴起［1］。ANN之所以受到人们的普遍关注，是由于它具有本质的非线形特征、并行处理能力、强鲁棒性以及自组织自学习的能力。其中研究得最为成熟的是误差的反传模型算法(BP算法，Back Propagation)，它的网络结构及算法直观、简单，在工业领域中应用较多。-Artificial Neural Networks (Aartificial Neural Network, under the short title of ANN) is a simulation of the structure of biological neurons and the information put forward a method of treatment. At early 1943, by Warren S. Mcculloch psychologists and mathematicians Walth H. Pitts proposed mathematical model of neurons, was left out in the cold for some time, and the rapid rise of the 80' s [1]. ANN because the general concern of the people is the essence of it is because of the non-linear characteristics of parallel processing capabilities, strong robustness and self-organizing capacity of self-learning. Research one of the most sophisticated are the error-propagation model of the algorithm (BP algorithm, Back Propagation), its network structure and algorithm for intuitive, simple, application in the industrial field more.
Platform: |
Size: 159744 |
Author: pjh |
Hits:
Description: 摘要图像特征的提取是视觉图像识别的重要方法之一，采用细胞神经网络（&’’）并行处理器进行图像特征的提取
具有实时快速的优点。该文将介绍&’’ 并行处理器的基本工作原理及其实现图像特征处理的逻辑组合通用方法，并以
图像的纹理分割与识别为例来说明&’’ 并行处理器应用于视觉图像识别的通用编程方法。-Abstract image feature extraction is a visual image to identify one of the important ways, the use of cellular neural networks for parallel processors (& ' ' ) image feature extraction with the advantages of rapid real-time. This paper will introduce & ' ' parallel processors to achieve the basic working principle and characteristics of image processing method of the logic combination of GM and image texture segmentation and recognition as an example to illustrate the application of parallel processors & ' ' visual image recognition General programming method.
Platform: |
Size: 1100800 |
Author: martin |
Hits:
Description: Based on the information processing functionalities of spiking neurons, a spiking neural network model is proposed to extract features from a visual image. The network is constructed with a conductance-based integrate-and-fire neuron model and a set of specific receptive fields. The properties of the network are detailed in this paper. Simulation results show that the network is able to perform image feature extraction within a time interval of 100 ms. This processing time is consistent with the human visual system. The demonstrations show how the network can extract right-angle contours in a visual image. Based on this principle, many other image features can be extracted by analogy. The parallel processing mechanism of this network model is very promising for a hardware implementation based on VLSI or FPGA technology.
Platform: |
Size: 288768 |
Author: sacoura31 |
Hits:
Description: 神经网络是单个并行处理元素的集合，我们从生物学神经系统得到启发。在自然界，网
络功能主要由神经节决定，我们可以通过改变连接点的权重来训练神经网络完成特定的功能-A single neural network is a collection of parallel processing elements, we have been inspired by biological nervous system. In nature, the network function of a decision by the ganglion, we can by changing the weight of the connection point to train neural network to complete the function of specific
Platform: |
Size: 244736 |
Author: zgp |
Hits:
Description: Parallel Processing, Neural Networks And Genetic Algorithms.djvu-Parallel Processing, Neural Networks And Genetic Algorithms
Platform: |
Size: 1746944 |
Author: bpodq |
Hits:
Description: 针对实际项目采用了webservice进行搭建分布应用及多线程进行分布式并行开发应用。本项目主要针对图象学的图象处理对已存在的图象进行数字识别，本项目还利用BP神经网络（人工智能）技术进行识别应用，本项目采用OCR、体征、BP三种识别方法进行有效识别，识别效率达到98 -100 ，运行效率95ms/56ms/0.9ms。-Webservice used for the actual project to build distributed applications and multi-threaded parallel development of distributed applications. The program is focused on image science image processing of already existing digital image recognition, the project also uses BP neural network (artificial intelligence) technology to identify applications, this project uses OCR, signs, BP three kinds of identification methods effective identification, identification efficiency of 98 -100 , operational efficiency 95ms/56ms/0.9ms.
Platform: |
Size: 12807168 |
Author: tony |
Hits:
Description: ：针对移动机器人传统路径规划算法效率不高，寻优能力差等问题，提出一种基
于神经网络和粒子群优化算法相结合的移动机器人路径规划方法．该方法利用神经网
络实现大量的并行和分布计算，发挥PSO简单、容易实现的优点，提高了路径规划的计
算效率和可靠性．仿真结果表明，这种新路径规划方法是可行且有效的．-The quality and eficiency of calculation is the two puzzling problems in the tradi—
tional algorithm for the robot path planning．In this paper，a new method of obstacle avoidance
and path planning based on neural network and particle swarm optimization is proposed．In
this method，a neural network is used to realize substantive parallel and distributed compu—
ting．And also this exerts the merit of PSO，which improves the computational eficiency and
reliability．As it is proved by analysis and test，that a better result is obtained by the pro—
posed algorithm．
Platform: |
Size: 162816 |
Author: 王风 |
Hits:
Description: 细胞神经网络(CNN)是一种和人类神经网络非常相似的并行计算模型，各个邻接节点间有不同的通信。在本程序中A模型是反馈矩阵，B是控制矩阵。-Cellular neural network (CNN) is very similar to the human neural network model of parallel computation, all adjacent nodes have different communication. A model of this process is the feedback matrix, B is the control matrix.
Platform: |
Size: 161792 |
Author: 严德春 |
Hits:
Description: 神经网络，它是一种模范动物神经网络行为特征，进行分布式并行信息处理的算法数学模型。这种网络依靠系统的复杂程度，通过调整内部大量节点之间相互连接的关系，从而达到处理信息的目的。-Neural network, it is a neural network model of animal behavior characteristics, the distributed parallel algorithm for information processing model. This network relies on the complexity of the system, by adjusting the connection between the number of nodes within the relationship, so as to achieve the purpose of processing information.
Platform: |
Size: 25600 |
Author: 秦莉平 |
Hits:
Description: 人工神经网络（ArtificialNeuralNetworks，简写为ANNs）也简称为神经网络（NNs）或称作连接模型（ConnectionistModel），它是一种模范动物神经网络行为特征，进行分布式并行信息处理的算法数学模型。这种网络依靠系统的复杂程度，通过调整内部大量节点之间相互连接的关系，从而达到处理信息的目的。
-Artificial neural network (ArtificialNeuralNetworks, abbreviated as ANNs) also referred to as neural networks (NNs) or known connection model (ConnectionistModel), it is a neural network model of animal behavior characteristics, the distributed parallel algorithm for information processing model. This network relies on the complexity of the system, by adjusting the connection between the number of nodes within the relationship, so as to achieve the purpose of processing information.
Platform: |
Size: 487424 |
Author: 小林 |
Hits:
Description: Multi-step-prediction of chaotic time series based
on co-evolutionary recurrent neural network
协同进化递归神经网络的多步混沌时间序列预测-This paper proposes a co-evolutionary recurrent neural network (CERNN) for the multi-step-prediction of chaotic
time series, it estimates the proper parameters of phase space reconstruction and optimizes the structure of recurrent
neural networks by co-evolutionary strategy. The searching space was separated into two subspaces and the individuals
are trained in a parallel computational procedure. It can dynamically combine the embedding method with the capability
of recurrent neural network to incorporate past experience due to internal recurrence. The eff ectiveness of CERNN is
evaluated by using three benchmark chaotic time series data sets: the Lorenz series, Mackey–Glass series and real-world
sun spot series. The simulation results show that CERNN improves the performances of multi-step-prediction of chaotic
time series.
Platform: |
Size: 152576 |
Author: |
Hits: