neural network and Matlab realized, with a lot of examples and I hope that you c...

用Java实现的多层BP神经网络，用于颜色和字符识别-using Java multi-storey BP neural network, for the color and character recognition

文中设计了一个3层径向基神经网络(RBFN)用于对企业的5项评价指标进行聚类分析，并与蚁群算法做了比较分析。RBFN由输入层 到隐含层采用传统的K一均值算法，隐含层到输出层通过“模2递减”学习速率的BP学习；蚁群算法根据信息素的分配能够自动调整收索 路径，从而达到数据自动聚类的目的。结果表明，与蚁群算法相比，改进RBFN具有快速收敛、自动识别奇异样本的优点，而蚁群算法 无须教师学习，并能够达到全局最优。-In this paper, we designed a 3-layer RBF neural network (RBFN) for the 5-to-business evaluation indicators cluster analysis and ant colony algorithm has done a comparative analysis. RBFN from input layer to hidden layer using the traditional K-means algorithm, hidden layer to output layer through the Mode 2 decreasing learning rate of BP learning ant colony algorithm based on pheromone can automatically adjust the allocation of land Faso path, thereby to achieve the purpose of automatic data clustering. The results showed that compared with the ant colony algorithm to improve the RBFN has a fast convergence, automatic identification of singular advantage of the sample, while the ant colony algorithm do not need teachers to learn and be able to reach the global optimum.

Hopfield Model Autoassociative Memory 源码, 经典的HOPFIELD人工神经网络例子源码-Hopfield Model Autoassociative Memory source, HOPFIELD classic example of artificial neural network source

提出了基于最小化估计的泛化误差界非优化估计方法。目前大多数的核参数选择方法都是通过极小化了来得到最优参数值，但是求解优化问题的计算代价相当的大，并且不能很好地体现数据的分布特征。本文采用非优化技术，通过极小化泛化误差来优化核及相关参数，由于直接计算最小半径和最大间隔，避免了对优化问题的直接求解，因此可以很好地降低计算代价。并且该方法直接从样本出发，可以很好地体现数据的分布特征，不管数据分布是否均匀都可以适用。给出了基于凸包估计的SVM核选择的模型及实现算法。 -Proposed based on minimizing the estimated generalization error bound of the non-optimal estimation method. Most of the nuclear parameter selection methods are to get the best by minimizing the parameter values, but the computational cost for solving optimization problems is quite large, and can not properly reflect the distribution characteristics of the data. In this paper, the non-optimization technology, by minimizing the generalization error to optimize the nuclear and related parameters, due to the direct calculation of the minimum radius and maximum interval, avoiding the direct solution of the optimization problem, it can very well reduce the computation cost. And that the method directly from the sample one can well reflect the distribution chara

该程序利用遗传算法求Rosenbrock函数的极大值。该算法主要包括确定决策变量、建立优化模型、确定编码方法、确定解码方法、确定个体评价方法、设计遗传算子、确定遗传算法的运行参数等七个步骤。-the procedures for the use of genetic algorithms Rosenbrock function of great value. The algorithm including the identification of the main decision variables, the optimal model to determine coding method to determine decoding, determine individual evaluation methods, genetic operator, genetic algorithm to determine the operating parameters seven steps.

遗传算法简称GA(Genetic Algorithm)，在本质上是一种不依赖具体问题的直接搜索方法。遗传算法在模式识别、神经网络、图像处理、机器学习、工业优化控制、自适应控制、生物科学、社会科学等方面都得到应用。在人工智能研究中，现在人们认为“遗传算法、自适应系统、细胞自动机、混沌理论与人工智能一样，都是对今后十年的计算技术有重大影响的关键技术”。-Referred to as genetic algorithm GA (Genetic Algorithm), in essence is a kind of do not rely on specific issues of direct search method. Genetic algorithm in pattern recognition, neural networks, image processing, machine learning, optimization of industrial control, adaptive control, biological sciences, social sciences and so on have been applied. Research in artificial intelligence, now that the " genetic algorithms, adaptive systems, cellular automata, chaos theory and artificial intelligence, are the next ten years of the computing technology have a major impact on the key technology."

人工股市（Artificial Stock Market，简称ASM）是模拟股市运作的一个程序。在这个电脑中的虚拟市场中，若干被称为交易者的人工智能程序(Agent)通过观察它们所在的数字世界中股价和股息的不断变换而做出预测，并且根据这些预测做出购买股票与否以及购买股票数量的决策。反过来，所有的交易者的决策又决定了股票的价格，这样，整个的股票交易市场就构成了一个自我封闭的计算系统。同时，这些交易者都具有学习的能力，可以根据以前预测的成功或者失败对自己的决策进行调整，并且通过一种被称为遗传算法的方法产生创新能力。-artificial stock market (Artificial Stock Market, or ASM) is a stock market simulation operation of the procedure. In this computer in the virtual market, a number of traders known as artificial intelligence (Agent) by observing them where the digital world and the stock dividend and the constantly changing projections, and based on these projections made whether or not to purchase stocks and the purchase of shares in decision-making. Conversely, all the decision-making transactions in turn determines the share price, so that the entire stock market would constitute a self-enclosed computer system. Moreover, these transactions are capable of learning, can be predicted based on previous success or failure of their decisions for the

bp神经网络计算的源程序-bp neural network calculation of the source

本人编写的QR分解的神经网络算法，该算法用QR分解神经网络的隐含层矩阵，可以作为学术上的比较和分析。适合各种benchmark问题，本人还加了自动产生分类向量和自动适应各种函数的例子-I prepared the QR decomposition of neural network algorithm using QR decomposition neural network hidden layer matrix, can be used as an academic comparison and analysis. Suitable for a variety of benchmark problems, I also added automatically generated classification and automatic vector function to adapt to a variety of examples

一个简单的遗传算法源代码。。。。。代码保证尽可能少，实际上也不必查错-A simple genetic algorithm source code. . . . . Code to ensure that as little as possible, in fact, do not have error checking