古典概型中样本空间选取教学法探讨

本文讨论了在古典概型教学中最重要,也是最基本的问题：样本空间的选取方法及在此问题中初学者易犯的各种错误与确定样本空间的正确思考方法及技巧。     

概率赋范线性空间的商空间

本文提出了M—PN空间的概率商空间的概念,证明了完备的M—PN空间的概率商空间是完备的M—PN空间。此外还证明了在M—PN空间中,概率致紧、概率列紧、概率预紧、概率可分性对其概率商空间具有遗传性。     

基于最小生成树与概率松弛结合的谱匹配算法

提出了一种基于最小生成树与概率松弛结合的谱匹配算法。该算法分别对给定的两个待匹配的特征点集构建最小生成树,通过最小生成树构造Laplace矩阵,由奇异值分解该矩阵得到的特征值和特征向量,计算出特征点匹配的初始概率,利用概率松弛迭代法,获得最终匹配结果。用大量的真实序列图像进行比较实验,结果验证了该算法的有效性和准确性。     

一种结合形状上下文分析的Laplace谱匹配算法

提出了一种结合形状上下文分析的Laplace谱匹配算法.工作主要侧重于如何提高Laplace谱匹配算法对点的位置随机抖动的鲁棒性.首先,使用Laplace矩阵的特征向量和特征值以及双随机矩阵的方法计算初始匹配概率.然后,借助于概率松弛算法,将用形状上下文表示的局部相似性融入Laplace谱匹配算法以优化谱匹配的结果.对真实和合成数据的实验表明该方法具有比较高的精度.     

具有区域极点约束的H∞概率鲁棒控制器设计与仿真

针对利用确定性方法对参数不确定系统进行具有区域椴点约束的H∞鲁棒控制综合具有计算复杂、保守性大等特点,一种基于概率鲁棒性分析思想的新控制器设计方法被提了出来.然后,利用多目标遗传算法对其进行求解.仿真结果显示,此设计方法简单、保守性小,而且能满足系统动态性能和对扰动抑制的鲁棒性要求,是有效、实用的.     

基于信息素评价的卫星数传调度蚁群算法

针对具有严格时间窗口和资源限制的复杂约束卫星数传调度问题,建立了基于蚁群算法的解构造图模型,提出了基于信息素局部评价方法、求和评价方法,相对评价方法的概率决策模型,设计了具有信息素全局更新策略的算法流程.仿真表明,基于不同信息素评价方法的蚁群算法能够取得较好效果,信息素相对评价方法在收敛速度和解寻优能力上都好于其他两种方法,而信息素求和评价方法相对较差.     

Inductive transfer learning for unlabeled target-domain via hybrid regularization

Recent years have witnessed an increasing interest in transfer learning. This paper deals with the classification problem that the target-domain with a different distribution from the source-domain is totally unlabeled, and aims to build an inductive model for unseen data. Firstly, we analyze the problem of class ratio drift in the previous work of transductive transfer learning, and propose to use a normalization method to move towards the desired class ratio. Furthermore, we develop a hybrid regularization framework for inductive transfer learning. It considers three factors, including the distribution geometry of the target-domain by manifold regularization, the entropy value of prediction probability by entropy regularization, and the class prior by expectation regularization. This framework is used to adapt the inductive model learnt from the source-domain to the target-domain. Finally, the experiments on the real-world text data show the effectiveness of our inductive method of transfer learning. Meanwhile, it can handle unseen test points.     

New computing method of weighted coefficients for tracking a maneuvering target using PDAF in the presence of clutter

To avoid or reduce the influence of unpredictable motion mode on data association, a new computing method of weighted coefficients of measurements for PDAF is presented in which it is assumed that the current turn rate of a maneuvering target changes within a limited range and its turn may be in arbitrary direction during data association. Thus, the predicted center for computing the weighted coefficients is a curved surface in 3-D space, which differs from the predicted center for setting up a validation gate, namely, a point in 3-D space. The distance between a measurement and the curved surface is used to compute its weighted coefficient. To reduce the computational complexity of weighted coefficients, the formulas for computing the maneuvering direction angle and turn rate corresponding to a measurement are presented. Simulation results show the proposed method reduces the percentage of lost tracks and improves the state estimation accuracy in tracking a maneuvering target using PDAF in the presence of clutter.     

基于全概率思想的排球发球技术水平的评价与分析

简要分析了传统的排球发球技术水平评价方法存在的不足,给出了排球发球质量等级划分办法,在此基础上,提出了排球发球技术水平的全概率评价方法.分别用2种方法对中国女排队员在北京奥运会上的发球技术水平进行了评价,并对2种评价结果进行了比较和分析,得出了全概率评价方法优于传统评价方法的结论.定义了发球威力的概念,计算了中国女排队员的发球威力值,并进行了2种评价方法评价结果与威力值的斯皮尔曼等级相关分析,得出全概率评价方法的评价结果与威力值具有较高一致性等结论.     

求解0-1动态优化问题的双概率原对偶遗传算法

在原对偶遗传算法(PDGA)的基础上,提出一种双概率原对偶遗传算法(DPPDGA).引入弱势基因位值与强势基因位值的概念,对二者赋予不同的对偶映射概率,并对两个对偶概率进行适应性调整.比较原始算法,改进算法使种群具有更理想的多样性,并利于种群较快地收敛到满意解.仿真结果表明,该算法在0-1动态优化问题的求解中具有更好的性能.