Description: 本算法在训练步数、训练时间及其误差精度等方面都优于常规的模糊神经网络，其学习收敛速度快、误差曲线也更稳定。-the algorithm steps in training, training time and error accuracy is superior to the conventional fuzzy neural network, learning fast convergence, the error curve is more stable. Platform: |
Size: 1024 |
Author:王鹏 |
Hits:
Description: 这是Armstrong的机器学习FAST算法matlab源代码，是从作者个人主页上下的，可以运行-This is Armstrong s Machine Learning FAST algorithm matlab source code from the author of the personal home page from top to bottom, you can run Platform: |
Size: 122880 |
Author:王寒涛 |
Hits:
Description: 超快的搜索算法，给学习数据结构的人一些启发-Ultra-fast search algorithm, data structure to the learning of people shed some light on the Platform: |
Size: 28672 |
Author:liangam |
Hits:
Description: 针对实际对象数学模型不明确而难以控制的问题,采用人工免疫网络的离散模
型与学习算法,将人工免疫系统与神经网络结构的优势相结合,提出了一种基于人工免疫
网络的模式识别算法,构造了对象识别的人工免疫网络模型.该算法综合了网络节点的定
位与参数调整以及对基函数的平滑因子实施调谐等功能,有效地解决了径向基函数
(RBF)神经网络模式识别的两个阶段任务,使模式识别的精度有较大的改进.采用两个不
同对象函数进行的仿真试验表明,该算法具有快速收敛性与较高的准确性.
-Mathematical model for the actual object is not clear and difficult to control the problem, the use of artificial immune network model and the discrete learning algorithm, artificial immune systems and neural networks combine the advantages of structure, an artificial immune network based on pattern recognition algorithms, Construction of the object identified by artificial immune network model. the algorithm is a combination of network nodes to adjust the position and parameters as well as the basis function implementation of the smoothing factor, tuning and other functions, to effectively solve the Radial Basis Function (RBF) neural network pattern recognition of two phase of the task, so that the accuracy of pattern recognition have greater improvements. the use of two different object functions of the simulation tests show that the algorithm has fast convergence and higher accuracy. Platform: |
Size: 114688 |
Author:杨玉琴 |
Hits:
Description: 人脸检测。用c++编写。人脸检测是人脸分析的首要环节，其处理的问题是确认图像（或影像）中是
否存在人脸，如果存在则对人脸进行定位。人脸检测的应用领域相当广泛，是实
现机器智能化的重要步骤之一。
AdaBoost 算法是1995 年提出的一种快速人脸检测算法，是人脸检测领域里
程碑式的进步，这种算法根据弱学习的反馈，适应性地调整假设的错误率，使在
效率不降低的情况下，检测正确率得到了很大的提高。-Face Detection. Using c++ to prepare. Face detection is the most important aspect of facial analysis, which dealt with the problem is to identify the image (or images) in the presence of human faces, if there is to locate human faces. Face detection applications of a wide range of intelligent machines to achieve one of the important steps. AdaBoost algorithm is proposed in 1995 a fast face detection algorithms, Face Detection is a landmark of progress in the field, the weak learning algorithm based on feedback, adaptive adjustment of the error rate assumptions, so that efficiency is not reduced in the case of , the detection accuracy has been greatly improved. Platform: |
Size: 897024 |
Author:ashura |
Hits:
Description: 爆炸万花筒屏幕保护
=== === === === === =====
一个纯代码的功能完整的爆炸万花筒屏幕保护程序，程序演示了可以爆炸很多随机彩色的几何状花朵。这个程序演示了一个完整的屏幕保护程序的编写过程，包括选项设置，可以设置花朵的浓度及爆炸的幅度，花朵的出现的多少，单薄，稀疏，花朵大小，移动速度快慢。支持预览模式参数设置和使用密码保护功能。程序使用 Polygon API函数绘制，速度很快。代码算法也比较经典，值得学习借鉴。-Kaleidoscope Screen Saver =================================== an explosion of pure explosion of the full functionality of the code kaleidoscope screen saver program, the program demonstrates that can blast a lot of random colorful geometric-shaped flowers. This program demonstrates a complete screen saver of the preparation process, including the option settings, you can set up flowers and explosive concentration of the magnitude of how much the emergence of flowers, thin, sparse, flower size, speed speed. Support the preview mode parameter setting and use the password protection. Program using the Polygon API function mapping, very fast. Code algorithm is also more classic, it is worth learning from. Platform: |
Size: 16384 |
Author:clinme |
Hits:
Description: 为了选择神经网络的最好结构以及增强模型的推广能力，提出一种自适应支持向量回归神经网络(SVR—NN)。SVR—NN 用支持向量回归(SVR)方法获得网络的初始结构和权值， 白适应地生
成网络隐层结点，然后用基于退火过程的鲁棒学习算法更新网络结点疹教和权 主。 SVR—NN有很
好的收敛性和鲁棒性，能抑制由于数据异常和参数选择不当所导致的“过拟合，’现象。将SVR—NN
应用到时间序列预测上。结果表明，SVR．NN预测模型能精确地预测混沌时间序列，具有很好的
理论和应用价值。-Abstract：To select the‘best’structure of the neural networks and enhance the generalization ability of models．a support
vector regression neural networks fSVR-NN)was proposed．Firstly，support vector regression approach was applied to
determine initial structure and initial weights of SVR．NN SO that the number of hidden layer nodes can be constructed
adaptively based on support vectors．Furthermore，an annealing robust learning algorithm was further presented to fine
tune the hidden node parameters and weights of SVR一ⅣM The adaptive SVR．NN has faSt convergence speed and robust
capability．and it can also suppress the ‘orerfitting’phenomena when the train data ncludes outliers．The adaptive
SVR．NN was then applied to time series prediction．Experimental results show that the adaptive SVR．ⅣⅣ can accurately
predict chaotic time series，and it iS valuable in both theory and application aspects． Platform: |
Size: 316416 |
Author:11 |
Hits:
Description: 该文件包含了BP小波神经网络快速学习算法研究与仿真的论文，以及两套小波算法方案，均通过MATLAB实现-This file contains the fast BP neural network learning algorithm and simulation papers, and two sets of wavelet algorithm for the program are achieved through the MATLAB Platform: |
Size: 148480 |
Author:蓉儿 |
Hits:
Description: 编程实现快速排序算法；实现文本文件拷贝函数copy.编写一个程序，要求：输入N个学生学号和成绩，并保存在stu.txt文本文件中，学生成绩进行排序并把排序后的结果输出到score.txt文件中,同时在屏幕上输出高于平均成绩的学生的学号和成绩。编写一个程序找出串str1和串str2的所有最长公共子串。
-Fast sorting algorithm programming text file copy function to achieve copy. Write a program that requires: Enter N-Number of students learning and achievement, and save the text file in the stu.txt, student achievement and to sort the results sorted output to score.txt file, while the output on the screen above the average score of the number of students learning and achievement. Write a program to identify strings str1 and str2 string of all longest common string. Platform: |
Size: 2048 |
Author: |
Hits:
Description: SVM is a supervised learning algorithm in the field of pattern recognition. latentsvm is fast version of it. Platform: |
Size: 63488 |
Author:selamicik |
Hits:
Description: 极限学习机，一种很快的学习算法，有很好的泛化性能，性能略低于SVM,但快于SVM。-Extreme learning machine, a very fast learning algorithm has good generalization of features, performance was slightly lower than SVM, but faster than the SVM.
Platform: |
Size: 3072 |
Author:zhuce1 |
Hits:
Description: 极限学习机的回归拟合及分类，介绍一个针对SLFN的新算法——极限学习机（Extreme Learning Machine，ELM），该算法随机产生输入层与隐含层间的连接权值及隐含层神经元的阈值，且在训练过程中无需调整，只需要设置隐含层神经元的个数，便可以获得唯一的最优解。与传统的训练方法相比，该方法具有学习速度快、泛化性能好等优点。-Extreme Learning Machine regression and classification, introduce a new algorithm for SLFN- Extreme Learning Machine (Extreme Learning Machine, ELM), the algorithm randomly generated input layer and hidden layer connection weights between the hidden layer and dollar threshold, and no adjustment in the training process, only need to set the number of hidden layer neurons, they can obtain a unique optimal solution. Compared with traditional training methods, this method has fast learning and good generalization performance. Platform: |
Size: 290816 |
Author:zhang mengyu |
Hits:
Description: “Fast Tracking via Dense Spatio-Temporal Context Learning,” In ECCV 2014的源代码，效果非常好。-In this paper, we present a simple yet fast and robust algorithm which exploits the spatio-temporal context for visual tracking. Our approach formulates the spatio-temporal relationships between the object of interest and its local context based on a Bayesian framework, which models the statistical correlation between the low-level features (i.e., image intensity and position) from the target and its surrounding regions. The tracking problem is posed by computing a confidence map, and obtaining the best target location by maximizing an object location likelihood function. The Fast Fourier Transform is adopted for fast learning and detection in this work. Implemented in MATLAB without code optimization, the proposed tracker runs at 350 frames per second on an i7 machine. Extensive experimental results show that the proposed algorithm performs favorably against state-of-the-art methods in terms of efficiency, accuracy and robustness. Platform: |
Size: 8955904 |
Author:happy |
Hits:
Description: 深度学习经典论文A fast learning algorithm for deep belief nets Why Does Unsupervised Pre-training Help Deep Learning等-classic paper for deep learning Platform: |
Size: 8092672 |
Author:惠开发 |
Hits:
Description: A fast learning algorithm for deep belief nets 文章代码-2006 A fast learning algorithm for deep belief nets Platform: |
Size: 18035712 |
Author:carole |
Hits:
Description: 稀疏编码算法是一种无监督学习方法，它用来寻找一组“超完备”基向量来更高效地表示样本数据-2006 A fast learning algorithm for deep belief nets Platform: |
Size: 10757120 |
Author:carole |
Hits:
Description: The main contribution of this paper is using
optimal control theory for improving the convergence
rate of backpropagation algorithm. In the proposed
approach, the learning algorithm of backpropagation
is modeled as a minimum time control problem
in which the step-size of its learning factor is considered
as the input of this model. In contrast to the traditional
backpropagation, learning algorithms which
the step-size by trial and error, it is selected
adaptively based on optimal control criterion. The effectiveness
of the proposed algorithm is uated in
two simulations: XOR and 3-bit parity. In both simulation
examples, the proposed algorithm outperforms
well in speed and the ability to escape local minima.-The main contribution of this paper is using
optimal control theory for improving the convergence
rate of backpropagation algorithm. In the proposed
approach, the learning algorithm of backpropagation
is modeled as a minimum time control problem
in which the step-size of its learning factor is considered
as the input of this model. In contrast to the traditional
backpropagation, learning algorithms which
the step-size by trial and error, it is selected
adaptively based on optimal control criterion. The effectiveness
of the proposed algorithm is uated in
two simulations: XOR and 3-bit parity. In both simulation
examples, the proposed algorithm outperforms
well in speed and the ability to escape local minima. Platform: |
Size: 415744 |
Author:samir |
Hits:
Search - fast learning algorithm