Sequential learning versus no learning in Bayesian regression models

We consider two regression models: linear and logistic. The dependent variable is observed periodically and in each period a Bayesian formulation is used to generate updated forecasts of the dependent variable as new data is observed. One would expect that including new data in the Bayesian updates results in improved forecasts over not including the new data. Our findings indicate that this is not always true. We show there exists a subset of the independent variable space that we call the “region of no learning.” If the independent variable values for a given period in the future are in this region, then the forecast does not change with any new data. Moreover, if the independent variable values are in a neighborhood of the region of no learning, then there may be little benefit to wait for the new data and update the forecast. We propose a statistical approach to characterize this neighborhood which we call the “region of little learning.” Our results provide insights into the trade‐offs that exist in situations when the decision maker has an incentive to make an early decision based on an early forecast versus waiting to make a later decision based on an updated forecast. © 2014 Wiley Periodicals, Inc. Naval Research Logistics 61: 532–548, 2014     

The sixteenth-century antecedents of special operations ‘small war’

The first conceptual, theoretical treatises about small war (la petite guerre) as special operations appeared only from the middle of the seventeenth century. The term is not used in the eighteenth-century sense of ‘special operations’ in older sources. The supposed absence of any treatment of the subject is surprising considering the obsession with the ‘art of war’ in the Renaissance, but other authors attribute it to a supposed antinomy between chivalric ideals and irregular warfare. But the absence of explicit manuals on the subject is not evidence of absence of advanced reflection on this kind of operations in the Middle Ages and in Early Modern times. We should thus look elsewhere, in other genres, for writings that contain and pass on military knowledge. Epics, romances, educational and military treatises, and memoirs in fact contain elements of a theory of special operations, even though these genres differ from our conception of rationality inherited from the Enlightenment.     

Guerrillas and bandits in the Serranía de Ronda, 1810–1812

The Spanish Guerrilla (1808–1812) which has given its name to ideologically motivated insurgencies is usually portrayed as a patriotic uprising against the French occupation forces of Napoleon. It was that, in part, but also many other things besides. This case study illustrates its overlap and convergence with banditry but also with social unrest turned into uprisings directed by poor Spaniards against their creditors, as in the storming of Ronda by insurgents in 1810. From the propaganda of the day to the subsequent Spanish patriotic historiography, there has been a tendency to exaggerate the amplitude of events and also the damage that was done to the French forces and the casualty figures inflicted on them.     

基于快速点特征直方图的特征点云迭代插值配准算法

为提高三维激光扫描点云数据的配准精度,提出了一种基于快速点特征直方图特征的迭代插值配准方法。配准过程中,点云数据获取时受扫描仪分辨率影响,点云局部或整体密度偏小,两次测量点云数据的相同位置不存在完全相同的点,以致对应点之间存在误差。为减小误差对配准精度影响,引入迭代插值方法,增加点云整体密度。通过计算关键点的快速点特征直方图的特征寻找对应关系,使用随机采样一致算法去除错误对应关系,对对应点协方差矩阵进行奇异值分解求得粗配准旋转平移矩阵,再使用迭代最近点算法进行点云的精确配准。实验结果表明,改进的配准方法简单、稳定可靠、计算速度有所增加,有效地提高了配准精度。     

弹药库防静电方法研究

根据弹药库静电危险场所的分类等级,以形成静电危害的3个基本条件为依据,讨论了弹药库防静电危害常用措施:设施设备静电接地、防静电工作台、人体静电防护、铺设导电(或防静电)地面以及各自防静电要求.这些措施为防止和消除弹药库的静电危害提供了根本保障.     

美国2014年《四年防务评估报告》特点

2014年3月4日,美国国防部向国会提交了2014年版《四年防务评估报告》和《2015财年国防预算报告》。《四年防务评估报告》更新了再平衡战略,为美军建设提出了优先发展方向,要求美军联合部队向“更小型化”转型,以满足未来军事行动需求。     

一类具有饱和发生率的离散型SIS传染病模型

研究一类具有饱和发生率的离散型SIS传染病模型,得到了模型的基本再生数.通过线性化的方法,运用LaSalle-Lyapunov定理,证明当基本再生数R01时,无病平衡点是全局渐近稳定的;通过迭代的方法,证明当基本再生数R01时,地方病平衡点是全局渐近稳定的.     

针对固定窗口算法的椭圆曲线密码故障分析

针对固定窗口算法实现点乘运算的椭圆曲线密码,基于符号变换故障攻击原理,通过分析不同故障模型下的密钥恢复过程,给出一种能够解决“零块失效”问题的改进故障分析方法,并进行仿真实验。实验结果表明：采用固定窗口算法的椭圆曲线密码易遭受故障攻击,10min内即可恢复NIST-192完整密钥。该故障分析方法也适用于其他采用点乘运算的密码算法。     

基于近似策略的联合优化方法

针对优化中收敛速度和优化解全局性的问题,提出了一种联合优化方法:构造原问题的近似模型,使用全局优化方法对近似函数进行优化,得到优化点作为初值,再使用局部优化方法对原问题进行优化.为了获得对原问题更好的近似,改进了径向基插值方法,以优化误差的方法来选择参数.利用临近空间机翼模型的优化对算法进行了测试,结果表明,优化参数的...     

SEA的侦察与监测系统效能分析

侦察与监测系统已成为现代C3I系统的重要组成部分,其效能评估既有C3I系统的一般性又有自身的特殊性。首先在一般系统效能含义上定义了侦察与监测系统的系统效能,比较了几种典型的C3I系统评估方法。在此基础上,将SEA方法应用于分析侦察与监测系统效能,并将情报信息量作为效能度量的指标。最后给出了分析计算过程和仿真结果。