搜索资源列表
Levenberg-Marquardt
- The LMA is a very popular curve-fitting algorithm used in many software applications for solving generic curve-fitting problems. However, as for many fitting algorithms, the LMA finds only a local minimum, which is not necessarily the global minimum.
multi-kernels-SVR
- 全局核(线性)与局部核(高斯)的加权组合,用于改进SVM的拟合和预测能力-The global kernel (linear) and local kernel (Gauss) weighted combination, to improve the SVM ability of fitting and prediction
PSO
- Rosenbrock函数优化属于无约束函数优化问题,其全局极小值位于一条平滑而狭长的抛物线形状的山谷底部,且为优化算法提供的信息很少,因此找到其全局极小值就显得很困难。根据Rosenbrock函数的这种特性,专门提出了一种改进的PSO算法,该算法引入三角函数因子,利用三角函数具有的周期振荡性,使每个粒子获得较强的振荡性,扩大每个粒子的搜索空间,引导粒子向全局极小值附近靠近,避免算法过早地收敛,陷入局部最优,从而找到Rosenbrock函数的全局极小值。大量实验结果表明,该算法具有很好的优化性能,
NGA
- 小生境遗传算法,该算法易于找出优化问题的所有局部最优解和全局最优解-Niche Genetic Algorithm. The algorithm is easy to find the optimization problem in all local optimal solution and the global optimal solution
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
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
hierachial_ensemble_of_global_local
- code for face identification by analysing local and global features
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
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.
POS_mod
- 改进的粒子群算法(PSO)MATLAB源程序m文件,在粒子群算法中引入克隆、选择算子寻求最优解。在同一粒子周围使用克隆选择算子进行多个方向的全局和局部搜索,促使种群中粒子快速进化,较快的得到局部最优和全局最优的位置-Improved particle swarm optimization algorithm (PSO) MATLAB source M files, in the particle swarm optimization algorithm to clone, the operat
quanjuliziqun
- 该程序为全局粒子群优化算法,进行优化计算时可避免陷入局部最优的陷阱。-The program for the global particle swarm optimization, optimization can be performed while avoiding local optimum trap.
DE-Clustering-Analysis
- 差分进化计算在解决这种聚类问题上表现出色,算法具有较强的通用性,不过分依赖于问题的信息;具有记忆个体最优解的能力、协同搜索的能力,以及可利用个体局部信息和群体全局信息指导算法进一步搜索的能力。-Differential evolutionary computation is excellent in solving this clustering problem. The algorithm has strong generality and can not depend on the info
dual-robot-path-planning
- 双机器人协调路径规划,局部路径使用蚁群算法,全局路径使用粒子群算法-Double coordinate path planning and local path using ant colony algorithm, the global path using the particle swarm optimization