搜索资源列表
minimize
- Conjugate Gradient Minimization在梯度下降算法中有着重要应用。可以解决一些一般方法不容易解决的问题
proj07-01
- 试验中使用模拟退火算法寻找一6-单元网络能量最小化的模型。模拟退火算法是模拟物理学上的退火技术。其优势在于有可能使系统从局部极小值跳出。-Simulated annealing algorithm used in the experiments to find a 6- unit network model of energy minimization. Simulated annealing algorithm simulates the physics of annealing. The a
webinar_files
- This a demonstration of how to find a minimum of a non-smooth objective function using the Genetic Algorithm (GA) function in the Genetic Algorithm and Direct Search Toolbox. Traditional derivative-based optimization methods, like those foun
EA_b
- 进化算法的一种 (1+1)EA的算法实现最优化求解 提供的示范函数为二进制数列求积累和-Opitmization method demo, based on (1+1)EA which belongs to Evolutionary Algorithm. An example function is provided as minimization of sum of a boolean array s bit multiplication.
svm
- 统计学习理论中提出的支撑向量机回归(SVR)遵循了结构风险最小化原则,从而避免了一味追求经验风险最小化带来的弊端-Statistical learning theory proposed by the support vector machine regression (SVR) to follow the structural risk minimization principle, thus avoiding the blind pursuit of Empirical Risk Minim
2DLDAwiththeSVM-basedfacerecognitionalgorithm
- 二维线性鉴别分析(2DLDA)算法能有效解决线性鉴别分析(LDA)算法的“小样本”效应，支持向量机 (SVM)具有结构风险最小化的特点，将两者结合起来用于人脸识别。首先，利用小波变换获取人脸图像的低频分量，忽 略高频分量：然后，用2DLDA算法提取人脸图像低频分量的线性鉴别特征，用“一对多”的SVM 多类分类算法完成人脸 识别。基于ORL人脸数据库和Yale人脸数据库的实验结果验证了2DLDA+SVM算法应用于人脸识别的有效性。-”Small sample size”problem
GA_method
- 直接使用实数编写的遗传算法，简单易行，不需要适应函数为正，可以计算含有约束的最小化问题-Prepared directly using the real genetic algorithm, is simple, do not meet the function is positive, can be calculated with constrained minimization problem
BPandHopfield
- 二次型最小化所展现的BP与Hopfield类型 神经网络的学习同质性，一篇不错的文章，pdf格式的。-Quadratic minimization demonstrated in BP and Hopfield type neural network learning homogeneity, a good article, pdf format.
Matlab-svm-BP-compare
- 支持向量机和BP神经网络虽然都可以用来做非线性回归，但它们所基于的理论基础不同，回归的机理也不相同。支持向量机基于结构风险最小化理论，普遍认为其泛化能力要比神经网络的强。为了验证这种观点，本文编写了支持向量机非线性回归的通用Matlab程序和基于神经网络工具箱的BP神经网络仿真模块，仿真结果证实，支持向量机做非线性回归不仅泛化能力强于BP网络，而且能避免神经网络的固有缺陷——训练结果不稳定。-SVM and BP neural networks, although non-linear regr
GENESIS
- 在TC和unix下用的，很好的一个遗传算法软件-GENESIS Version 5.0 was written to promote the study of genetic algorithms for function minimization
maxflow-v3.03.src
- This software library implements the maxflow algorithm described in An Experimental Comparison of Min-Cut/Max-Flow Algorithms for Energy Minimization in Vision. Yuri Boykov and Vladimir Kolmogorov. In IEEE Transactions on Pattern Analys
WCA
- 一种新的智能优化算法，水循环算法，可以用来求解优化问题-water cycle. This code is prepared for single objective function (minimization), unconstrained, and continuous problems.
StatLSSVM
- 支持向量机工具箱By Kris De Brabanter,标准的非参数回归，健壮的回归，一些调优标准等经典交叉验证，较好的交互性-The StatLSSVM toolbox is written so that only a few lines of code are necessary in order to perform standard nonparametric regression, regression with correlated errors and robust regre