登录
首页 » Matlab » 证券及期货条例: Submodular 函数优化工具箱

证券及期货条例: Submodular 函数优化工具箱

于 2022-02-13 发布 文件大小:297.24 kB
0 79
下载积分: 2 下载次数: 1

代码说明:

    近些年,在机器学习(Machine Learning)的各领域里,出现了一个非常有用的基本问题结构:亚模性(子模性,次模性)(submodularity)是一个直觉上收益递减(diminishing return)的性质,声明给一个小集合增加元素比给一个大集合增加同一个元素更有帮助。与凸性(convexity)类似,亚模性可以有效找出大规模问题的可证明的(近似)最优解。    SFO:一个MATLAB/Octave工具箱,实现各种用于亚模函数最小化/最大化的算法。   相关论文发表在Journal of Machine Learning  Research(JMLR)11(2010)1141-1144   【注】本人已将该论文添加在源代码包之中。     希望您使用愉快!^-^ 

下载说明:请别用迅雷下载,失败请重下,重下不扣分!

发表评论

0 个回复

  • gps_final_new
    说明:  gps卫星位置预报,MATLAB编写,有动画界面(gps satellite position prediction, MATLAB prepared)
    2010-04-02 11:01:59下载
    积分:1
  • obstacles
    一个在matlab下编写的关于机器人路径规划的小程序(A matlab prepared on a small robot path planning process)
    2010-05-20 09:42:27下载
    积分:1
  • ejcf
    该程序可用于近红外光谱二阶求导预处理,经本人实测,程序可用。(The program can be used near-infrared spectroscopy second derivative pretreatment, as I found, the program is available.)
    2014-05-15 14:20:53下载
    积分:1
  • cp0801_PAMreceiver
    超宽带脉冲幅度调制接收端频谱,与其他的几个共用(Ultra-wideband pulse amplitude modulation spectrum receiver, and some other common)
    2009-03-26 20:22:10下载
    积分:1
  • C45_Sun
    常用决策树算法C4.5的实现代码。利用matlab实现。(C4.5 decision tree algorithm used in the realization of the code. Realize the use of matlab.)
    2008-04-14 21:55:20下载
    积分:1
  • lorenzsys
    lorenz系统simulink仿真程序,初学者入门程序(lorenz system simulink simulation program, beginners entry procedures)
    2013-07-11 16:32:13下载
    积分:1
  • paper
    USING MATLAB TO CREATE AN IMAGE FROM RADAR
    2011-11-24 20:17:44下载
    积分:1
  • QRSTutorial
    The Principles of Software QRS Detection
    2010-05-17 17:07:04下载
    积分:1
  • DSP_RADAR_SYSTEM_DESIGN
    主要用于实现雷达信号的处理,对四目标进行识别测速(Mainly used to implement the processing of radar signals for the four target recognition speed)
    2014-12-23 09:53:46下载
    积分:1
  • mutli-output
    多输出的支持向量机matlab代码, 常规的svm是单输出,这个是多输出( Standard SVR formulation only considers the single-output problem. In the case of several output variables, other methods (neural networks, kernel ridge regression) must be deployed, but the good properties of SVR are lost: hinge-loss function and sparsity. The proposed model M-SVR extends the single-output SVR by taking into account the nonlinear relations between features but also among the output variables, which are typically inter-dependent. )
    2016-03-05 15:29:54下载
    积分:1
  • 696516资源总数
  • 106914会员总数
  • 0今日下载