搜索资源列表
SA
- 模拟退火算法的MATLAB实现,通过赋予搜索过程一种时变且最终趋于零的概率突跳性,从而可有效避免陷入局部极小并最终趋于全局最优的串行结构的优化算法。-MATLAB implementation of simulated annealing algorithm, a search process by giving time-varying and ultimately tends to zero the probability of jumps, which can effectively av
11
- RBF网络的学习过程与BP网络的学习过程类似,两者的主要区别在于各使用不同的作用函数。BP网络中隐层使用的是Sigmoid函数,其值在输入空间中无限大的范围内为非零值,因而是一种全局逼近的神经网络;而RBF网络中的作用函数是高斯基函数,其值在输入空间中有限范围内为非零值,因为RBF网络是局部逼近的神经网络。 RBF网络是一种3层前向网络,由输入到输出的映射是非线性的,而隐层空间到输出空间的映射是线性的,而且RBF网络局部逼近的神经网络,因而采用RBF网络大大加快学习速度并避免局部极小问题,
Files
- I send a file that contains 7 source code and simulation in matlab. source codes contain GA and HGAPSO and PSO local&global.simulink files contain svpwm,matrix converter direct&improve.
Genetic-Algorithm-matlab
- 遗传算法 ( Genetic Algorithm , GA) 是借鉴生物界自然选择和群体进化机制形成的一种全局寻优算法 。与传统的优化算法相比 ,遗传算法具有如下优点 [1 ] :1 ) 不是从单个点 ,而是从多个点构成的群体开始搜索 2) 在搜索最优解过程中 ,只需要由目标函数值转换得来的适应值信息 ,而不需要导数等其它辅助信息 3) 搜索过程不易陷入局部最优点 。 数学建模中常用的matlab算法,遗传算法,内容详细,包括PDF版本的详细的算法实现过程;-Genetic Algorith
beibao
- 模拟退火算法是一种通用的随机搜索算法,是对局部搜索算法的扩展。与一般局部搜索算法不同,SA以一定的概率选择领域中目标值相对较小的状态,是一种理论上的全局最优算法。-Simulated annealing algorithm is a common random search algorithm is an extension of local search algorithm. Different general local search algorithm, SA with a certain
SAPSO
- 本程序介绍了一种改进的粒子群寻优算法;该算法可以更好的实现粒子群寻优过程中的全局搜索与局部搜索值间的平衡。从而寻得最优结果。-This program introduces an improved particle swarm optimization algorithm This algorithm can better realize the particle swarm optimization in the process of the balance between global se
Simulated-annealing-algorithm
- 模拟退火算法,是通过赋予搜索过程一种时变且最终趋于零的概率突跳性,从而可有效避免陷入局部极小并最终趋于全局最优的串行结构的优化算法。-Simulated annealing algorithm is a time-varying and ultimately approach zero probability of sudden rebound, which can effectively avoid the local minimum by giving the search process
project
- The eight-node quadrilateral element is a two-dimensional finite element with both local and global coordinates
CPSO
- 混沌粒子群算法,求代价函数问题,可以避免粒子群算法陷入局部最优,求取全局最优。代码含代价函数,可作为例子理解。-Chaos particle swarm optimization, seeking cost function problems, avoid getting into local optimum particle swarm algorithm, obtaining the global optimum. Code containing cost function can be u
hierachial_ensemble_of_global_local
- code for face identification by analysing local and global features
vns-mpso
- 将NSGA-II 的理念融入粒子群中,并加入保存优秀解的机制,提出MPSO算法。针对MPSO 易陷入局部极值的缺点,加入变邻域机制,通过随机的破坏旧解和重建新解,找到全局最优解,以达到帮助粒子跳出局部极值的目的,提 出VNS-MPSO 算法。-Will the NSGA- II concept into particle swarm, and join save good solution mechanism, MPSO algorithm is put forward. Against d
ABC
- 人工蜂群算法是模仿蜜蜂行为提出的一种优化方法,只需要对问题进行优劣的比较,通过各人工蜂个体的局部寻优行为,最终在群体中使全局最优值突现出来,有着较快的收敛速度。-Artificial colony algorithm is to imitate the bee behavior put forward an optimization method,Only need to compare advantages and disadvantages of problem, Through the l
The-Cuckoo-SearchThe-Cuckoo-Search
- 布谷鸟搜索(CS)算法是根据生物界中布谷鸟的寄生繁殖机理而提出的一种仿生智能优化算法,由于布谷鸟搜索算法具有优秀的全局搜索和局部搜索能力,并且控制参数少,收敛速度快-The Cuckoo Search (CS) algorithm is a bionic intelligent optimization algorithmbased on the mechanism of biological reproduction in parasitic cuckoo proposed. Dueto th
altificial-bee-colonyalgorithm
- 人工蜂群算法是模仿蜜蜂行为提出的一种优化方法,是集群智能思想的一个具体应用,它的主要特点是不需要了解问题的特殊信息,只需要对问题进行优劣的比较,通过各人工蜂个体的局部寻优行为,最终在群体中使全局最优值突现出来,有着较快的收敛速度。-Artificial bee colony algorithm is an optimization technique to mimic the behavior of bees make is thinking of a particular cluster of
NSGA-II
- NSGA-II非支配排序遗传算法,可避免陷入局部最优,属于全局算法-NSGA-II non dominated sorting genetic algorithm, can avoid falling into local optimal, belong to the global algorithm
Matlab_Code_TAFGLSE_ISCIDE_2013
- Texture-Aware Fast Global Level Set Evolution-Due to its intrinsic advantages such as the ability to automatically handle complex shapes and topological changes, the level set method has been widely used in image segmentation. Nevertheless, in additi
GAKMeans
- 由于Kmeans聚类分析是一个局部的搜索过程,因此加入遗传算法进行全局搜索选择最优的初始中心点使得Kmeans算法产生较大的改进-Since Kmeans Cluster analysis is a local search process, so join a global search for the genetic algorithm to the optimal initial centers such Kmeans algorithm produces greater improve
PSOPT_Manual_R3
- PSOPT是以C++编写的开源的最优控制软件包,它采用直接配置法。这些方法求解最优控制问题,是使用全局或局部多项式来逼近随时间变化的各种变量,这就可以将微分方程和连续约束离散化,并可用众所周知的数值求积公式来计算与最优控制问题有关的任何积分,然后,非线性规划就可被用于求解局部最优解。-PSOPT is an open source optimal control package written in C++ that uses direct collocation methods. Thes
nef-1.4.0
- 非线性滤波框架(nef),包括了 EKF,UKF,DDF1 DDF2,CDF,迭代滤波器,随机积分滤波器, 组合滤波器, 集合卡尔曼滤波, 高斯和滤波,粒子滤波,自回归最小二乘方法-nonlinear estimation framework (NEF) toolbox A. Implemeted local estimation techniques: a1. (extended) Kalman filter a2. Unscented Kalman filter a3.
disparity_postorocessing
- 立体匹配中,不管是局部还是全局匹配方法得到的视差图,需要对视差图进行后处理,比如左右一致性检验,遮挡过滤,中值滤波。-Stereo matching, whether local or global matching disparity map obtained after processing the need for the disparity map, such as about consistency checking, blocking filter, median filter.