量化子空间分布隐马尔可夫模型的间接训练

研究了量化子空间分布隐马尔可夫模型(quantized subspace distribution hidden markov model，QSDHMM)间接训练所涉及的三个关键的问题：连续分布隐马尔可夫模型(continuous distribution hidden markov model，CDHMM)的训练、特征子空间的划分和子空间高斯分布的量化方案。提出了相关特征子空间的定义及实现算法。在仿真实验中，采用基于分类学的：Bhattacharyya距离测度的K均值高斯聚类算法对子空间高斯分布进行量化，用最相关子空间的概念划分特征子空间，并将这两者结合使用，提高了系统的识别精度和速度。     

The calculation of static polarizabilities of 1-3D periodic compounds. the implementation in the crystal code

The Coupled Perturbed Hartree-Fock (CPHF) scheme has been implemented in the CRYSTAL06 program, that uses a gaussian type basis set, for systems periodic in 1D (polymers), 2D (slabs), 3D (crystals) and, as a limiting case, 0D (molecules), which enables comparison with molecular codes. CPHF is applied to the calculation of the polarizability alpha of LiF in different aggregation states: finite and infinite chains, slabs, and cubic crystal. Correctness of the computational scheme for the various dimensionalities and its numerical efficiency are confirmed by the correct trend of alpha: alpha for a finite linear chain containing N LiF units with large N tends to the value for the infinite chain, N parallel chains give the slab value when N is sufficiently large, and N superimposed slabs tend to the bulk value. CPHF results compare well with those obtained with a saw-tooth potential approach, previously implemented in CRYSTAL. High numerical accuracy can easily be achieved at relatively low cost, with the same kind of dependence on the computational parameters as for the SCF cycle. Overall, the cost of one component of the dielectric tensor is roughly the same as for the SCF cycle, and it is dominated by the calculation of two-electron four-center integrals.     

谐振子势与高斯势的联合势阱中两分量旋转玻色爱因斯坦凝聚体的基态特性和自旋纹理

研究了两分量旋转玻色爱因斯坦凝聚体在谐振子势与高斯势的联合势阱中的基态特性和自旋纹理。通过托马斯-费米近似得到每组分凝聚体在相混合态时密度分布首次形成中心洞的临界旋转角频率,并根据旋转角频率与临界旋转角频率的关系,给出了两分量凝聚体的三种不同的基态密度分布:两个都是盘、一个是盘和另一个是环、两个都是环。对于相分离的情况,针对两分量粒子数严重不平衡的凝聚体分别作托马斯费米近似,解析地给出了两分量凝聚体的两种对称基态密度分布。同时研究了凝聚体在两分量的界面处形成的两种赝自旋纹理,它们分别是巨斯格明子和同轴双环斯格明子。     

基于混合变异的果蝇优化算法

果蝇算法是新提出一种的群智能优化算法,它存在一些不足尤其是在收敛性和求解精度方面.基于以上提出了一种基于混合变异算子的果蝇优化算法,充分利用柯西变异算子所具有全局搜索能力强的特点和高斯变异算子的局部搜能力强的优点,将这两个算子结合在一起来更新果蝇的位置从而很好的避免了各自算子的缺点.为了验证算法的性能通过测试7个标准多元非线性函数同果蝇优化算法及参考文献中算法结果相比较,实验表明该算法的收敛速度和求解的精度都得到了提高.     

基于自适应混合高斯模型的前景目标检测

在计算机视觉研究中,从视频序列中提取出前景目标是关键步骤之一。而混合高斯背景模型是前景目标检测的一种常用算法。针对传统混合高斯建模过程中分别对每个像素建立固定个数的高斯模型和相同的学习率这一缺陷,本文先对视频帧进行了分块处理,然后自适应的对每个像素块采取不同的高斯分布个数和学习率,并且在建模过程的不同时间段采用不同的学习率,最后对检测结果在空域上进行数学形态学的处理。实验结果表明,与传统检测方法相比,该方法能够更加准确和快速地检测出前景目标。     

Enhanced Raman Scattering of Elliptical Laser beam in a Collisional Plasma

This paper presents the Enhanced Raman scattering of a elliptical laser beam in a collisional plasma. We have considered the mechanism of non‐uniform heating of carriers along the wave‐front, which is important in collisional plasma. The nonlinearity arising through non‐uniform heating leads to redistribution of carriers, which modifies the background plasma density profile in a direction transverse to pump beam axis. This modification in density effects the incident laser beam, plasma wave and back‐scattered beam. Non‐linear differential equations for the beam width parameters of pump laser beam, plasma wave and back‐scattered beam are set up and solved numerically. Numerical results predict the effect of self‐focusing of waves on the back‐scattered beam (© 2011 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

基于协作机器人仿真环境的关节轨迹预测方法

安全在人机协作过程中是至关重要的,必须实时掌握人的行为信息,并进行准确高效的预测。本文基于Linux和ROS系统搭建仿真环境,通过Xtion PRO LIVE深度相机采集多组人体关节的空间位置信息,然后通过无监督学习方法对采集到的坐标点进行聚类和预测,实时更新预测模型,并基于minimum-jerk对特殊异常轨迹进行预测。为了充分保证人的安全,主要研究手部和肘部运动轨迹的预测方法。最终实验结果证明,本文所提出的分层轨迹预测框架可以很好的描述人体运动轨迹,并实时做出准确的预测,不仅保证了人体安全,而且对于提高生产效率具有重要意义。     

改进的邻域支持向量解算法

针对实施邻域风险最小化原则的邻域支持向量解算法,根据被错分样本一定是支持向量提出一种利用支持向量删除训练样本中难学习样本的修剪算法;依据最大似然原则对已有的高斯邻域函数参数取值方法进行改进.初步实验表明,训练样本的修剪与邻域函数参数取值方法的改进可明显提高邻域支持向量解算法的泛化能力,比SVM测试准确率提高0.5%左右.     

Accuracy and convergence of free energy differences calculated from nonequilibrium switching processes

The molecular-dynamics-based calculation of accurate free energy differences for biomolecular systems is a challenging task. Accordingly, convergence and accuracy of established equilibrium methods has been subject of many studies, often focusing at small test systems. In contrast, the potential of more recently proposed nonequilibrium methods, derived from the Jarzynski and Crooks equalities, has not yet fully been explored. Here, we compare the performance of these methods by calculating free energy differences for test systems at different levels of complexity and varying extent of the involved perturbations. We consider the interconversion of ethane into methanol, the switching of a tryptophane-sidechain in a tripeptide, and the binding of two different ligands to the globular protein snurportin 1. On the basis of our results, we suggest and assess a new nonequilibrium free energy method, Crooks Gaussian Intersection (CGI), which combines the advantages of existing methods. CGI is highly parallelizable and, for the test systems considered here, is shown to outperform the other studied equilibrium and nonequilibrium methods.     

Multi‐scale Gaussian/Haar wavelet strategies coupled with sub‐window factor analysis for an accurate alignment in nontargeted metabolic profiling to enhance herbal origin discrimination capability

Metabolic dataset can provide an overview of different herbal origin, which is conducted by some statistical procedures. Such results often deviate to a certain degree, due to peaks shifts in chromatographic signals. In order to solve this problem, an improved algorithm of combining sub‐window factor analysis with the mass spectrum information is proposed. The algorithm uses a peak detection approach derived either from multi‐scale Gaussian function or Haar wavelet to locate the peaks with different application scope; the candidate drift points at each peak are estimated by Fast Fourier transform cross correlation; Specifically, the best drift points at each candidate peaks are confirmed by sub‐window factor analysis and mass spectrum information in nontargeted metabolic profiling. Finally, the peak regions were aligned against a reference chromatogram, and the non‐peak regions were used linear interpolation. The chromatographic signals of 30 Bupleurum samples were aligned as an illustration of this algorithm, and they could be well distinguished using some statistical procedures. The result demonstrates that the presented method is stronger than other mass‐spectra based algorithms, when facing the alignment of some co‐eluted peaks.