CPN算法的matlab实现，CPN人工神经网络

Matlab fuzzy control progrom.

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

遗传算法求解优化问题，简明教程加示例- The heredity algorithm solution optimization question, the concise course adds the demonstration

仿真1：首先把网络温度参数T固定在100，按工作规则共进行1000次状态更新,把这1000次状态转移中网络中的各个状态出现的次数Si（i＝1，2，…，16）记录下来 按下式计算各个状态出现的实际频率： Pi=Si/∑i=1,NSi=Si/M 同时按照Bo1tzmann分布计算网络各个状态出现概率的理论值： Q（Ei）=（1/Z）exp（-Ei/T） 仿真2：实施降温方案，重新计算 采用快速降温方案：T（t）= T0/（1+t） T从1000降到0.01，按工作规则更新网络状态 当T=0.01时结束降温，再让T保持在0.01进行1000次状态转移，比较两种概率-a simulation : First of all network parameters temperature T fixed at 100 and, according to the rules for a total of 1000 to update the state, this state of the 1000 network transfer of all states for the number of Si (i = 1, 2, ..., 16) all recorded determined by the formula state-of the actual frequency : Pi = Si/i = 1, NSi = Si/M in accordance with Bo1tzmann distributed computing network of states all probability the theoretical value : Q (Ei) = (1/Z) exp (- Ei/T) Simulation 2 : implementation of cooling, re-using rapid cooling programs : T (t) = T0/(1 t) T dropped to 0.01 from 1000 and, according to the rules updated network state when T = 0.01 at the end of cooling, let T at 0.01 for the 1000 state tran

JAVA版的蚂蚁算法(Ant Colony Optimization Algorithms) -JAVA version of the ant algorithm (Ant Colony Optimization Algorithms)

本程序利用蚂蚁算法解决TSP（旅行商问题）问题-This procedure using ant algorithm to solve TSP (Traveling Salesman Problem) problem

Demonstration and test of the kernel-ica package

遗传算法解决随机期望值模型问题的例题-random genetic algorithms to solve the problem expectations model example

遗传算法求解车辆路径问题，没有解压缩的密码。-Genetic Algorithm for vehicle routing problem, no extract passwords.

基于遗传算法的模糊卡尔曼滤波器-Fuzzy kalman filter based on inheritance algorithmic.