Classification Using Generalized Partial Least Squares

Advances in computational biology have made simultaneous monitoring of thousands of features possible. The high throughput technologies not only bring about a much richer information context in which to study various aspects of gene function, but they also present the challenge of analyzing data with a large number of covariates and few samples. As an integral part of machine learning, classification of samples into two or more categories is almost always of interest to scientists. We address the question of classification in this setting by extending partial least squares (PLS), a popular dimension reduction tool in chemometrics, in the context of generalized linear regression, based on a previous approach, iteratively reweighted partial least squares, that is, IRWPLS. We compare our results with two-stage PLS and with other classifiers. We show that by phrasing the problem in a generalized linear model setting and by applying Firth's procedure to avoid (quasi)separation, we often get lower classification error rates.     

Investigation of 2D-photonic crystal resonant cavity based WDM demultiplexer

In this attempt, Two Dimensional Photonic Crystal (2DPC) Quasi Square Ring Resonator (QSRR) based four channel demultiplexer is proposed and designed for Wavelength Division Multiplexing systems. The performance parameters of the demultiplexer such as transmission efficiency, passband width, line spacing, Q factor and crosstalk are investigated. The proposed demultiplexer is composed of bus waveguide, drop waveguide and QSRR. In the proposed demultiplexer, the output ports are arranged separately in odd and even number, where an odd number of ports are located on the right side and even number of ports are located on the left side of the bus waveguide that are used to reduce the channel interference or crosstalk. Further, the refractive index of rods around the center rod is increased linearly one to another in order to improve the signal quality. The resonant wavelengths of the proposed demultiplexer are of 1521.1?nm, 1522.0?nm, 1523.2?nm and 1524.3?nm, respectively. The footprint of the device is of 180.96?μm². Then, a four channel point to point network is designed and the proposed four channel demultiplexer is implemented by replacing a conventional demultiplexer. Finally, functional parameters of the network, namely, BER, receiver sensitivity and Q factor are estimated by varying the link distance. This attempt could create new dimensions of research in the domain of photonic networks.     

准分子CdNe的相对论赝势SCF/SDCI计算

用相对论赝势从头算自洽场/单双重电子激发电子相关方法(SCF/SDCI)对准分子CdNe的基态(X1Σ+)和几个低激发态(A3Π、B3Σ、C1Π和D1Σ+)进行了计算,得到了这些态的电子结构、势能曲线及光谱常数,并进行了讨论     

集值映射的伪(*)连续与弱(*)连续性

本文引入了集值映射的伪（＊）连续性与弱（＊）连续性概念的定义,研究了伪（＊）连续、（＊）连续及弱（＊）连续的等价关系,最后研究了乘积空间中的集值映射成为伪（＊）连续和弱（＊）连续的充要条件。     

射频放电CO2激光器放电区物理过程的研究

本文认为射频放电CO₂激光器的高增益区是类法拉第暗区,并利用有质动力计算了阴极区和类法拉第暗区电子的能量及类法拉第暗区的尺寸,最后给出了四极放电CO₂激光器的一些实验结果。     

方形单元衍射光学阵列器件实现近圆形均匀焦斑

本文以方形激光输入、圆形均匀焦斑输出为例,分析比较了整体式和阵列式衍射光学器件获取圆形均匀焦斑的能力.通过在阵列式器件方形单元间附加线性相移来改变各子焦斑的相互位置、重叠区域,从而逼近于一定直径的近圆形均匀焦斑.该方法将二维问题转化为一维,减少计算量.采用基于爬山法与模拟退火法相结合的混合算法设计相位分布.结果表明了该方法的有效性,并具有获得任意形状均匀焦斑的灵活性。     

拟二次系统的广义焦点量与极限环分枝

本文给出了拟二次系统的前18个奇点量和可积性条件,由此统一解决了几类实平面微分自治系统的初奇点、高次奇点以及无穷远点的中心焦点判定与极限环分枝问题.     

用类氖锗X光激光实验检验理论模拟

 介绍了用国内1ns长脉冲和国外多脉冲驱动的类氖锗的实验数据检验X光激光的理论模拟程序系列的工作。在长脉冲驱动时，对波长19.6，23.2和23.6nm三条激光线，理论模拟与实验数据基本符合。在多个短脉冲驱动时，除了与严重过电离的复合等离子体有关的实验外，理论模拟与实验也基本符合。为今后设计激光聚变等离子体诊断所需类氖锗X光激光光源工作打下了基础。     

MAXIMUM INFORMATION AND OPTIMUM ESTIMATING FUNCTION

In order to construct estimating functions in some parametric models, this paper introduces two classes of information matrices. Some necessary and sufficient conditions for the information matrices achieving their upper bounds are given. For the problem of estimating the median, some optimum estimating functions based on the information matrices are acquired. Under some regularity conditions, an approach to carrying out the best basis function is introduced. In nonlinear regression models, an optimum estimating function based on the information matrices is obtained. Some examples are given to illustrate the results. Finally, the concept of optimum estimating function and the methods of constructing optimum estimating function are developed in more general statistical models.     

Every 4-connected line graph of a quasi claw-free graph is hamiltonian connected

Let G be a graph. For u,vV(G) with dist_G(u,v)=2, denote J_G(u,v)={wN_G(u)∩N_G(v)|N_G(w)N_G(u)N_G(v){u,v}}. A graph G is called quasi claw-free if J_G(u,v)≠ for any u,vV(G) with dist_G(u,v)=2. In 1986, Thomassen conjectured that every 4-connected line graph is hamiltonian. In this paper we show that every 4-connected line graph of a quasi claw-free graph is hamiltonian connected.