Sparse Multicategory Generalized Distance Weighted Discrimination in Ultra-High Dimensions

Distance weighted discrimination (DWD) is an appealing classification method that is capable of overcoming data piling problems in high-dimensional settings. Especially when various sparsity structures are assumed in these settings, variable selection in multicategory classification poses great challenges. In this paper, we propose a multicategory generalized DWD (MgDWD) method that maintains intrinsic variable group structures during selection using a sparse group lasso penalty. Theoretically, we derive minimizer uniqueness for the penalized MgDWD loss function and consistency properties for the proposed classifier. We further develop an efficient algorithm based on the proximal operator to solve the optimization problem. The performance of MgDWD is evaluated using finite sample simulations and miRNA data from an HIV study.     

蓝宝石冲击消光晶向效应的第一性原理

蓝宝石的冲击消光现象是高压领域中的研究热点.低压段(86 GPa范围内)的实验研究表明蓝宝石的冲击消光与晶向相关,但在高压段(压力范围:131255 GPa)是否也具有晶向相关性目前尚不清楚.为此,利用第一性原理方法,分别计算了八个不同晶向的蓝宝石理想晶体和含氧离子空位缺陷晶体在高压段的光吸收性质,结果发现:1)蓝宝石在高压段的冲击消光表现出明显的晶向效应,且该效应还随压力增大而增强;一步的数据分析可以看出,在冲击实验采用的波段内,a晶向的消光最弱(透明性最好),c晶向的消光最强与c晶向的消光接近,g晶向的消光要弱于s晶向的消光.鉴于此,如果在高压段开展加窗冲击波实验,建议选择a晶向或m晶向的蓝宝石作为其光学窗口.本文结果不仅有助于深入地认识蓝宝石在极端条件下的光学性质,而且对未来的实验研究有重要的参考作用.     

光栅衍射实验中测量光栅常数的两种辅助方法

衍射光栅可以将不同波长的光在空间上分离开,在光谱分析领域中有广泛应用.利用光栅的衍射来测量光栅常数是大学物理实验中的一个典型实验,而本文介绍的正是光栅衍射实验中的两个有趣现象:一是旋转光栅时,谱线的偏转角存在一个最小值;二是光栅旋转到一定角度时,衍射光斑会逐渐消失在视野中.论文从理论上分析解释这两种现象,并设计出两种测量光栅常数的方法,在实验上进行验证.与传统的用最小偏转角测光栅常数的方法相比,本文重点分析了由判定最小偏转角位置偏差所带来的影响,并提出了对称旋转测量的方法来进一步减小误差.     

Investigation of electrochemical performance of MgNiO2 prepared by sol-gel synthesis route for aqueous-based supercapacitor application

In this work, we have successfully synthesized MgNiO₂ using a sol-gel wet chemical synthesis technique named MNO - 3. Electrochemical measurements in the presence of aqueous 1 M Li₂SO₄ electrolyte indicate that MNO - 3 samples exhibit a capacitance value of about 30 F/g and an energy density of about 20 Wh/kg. Subsequently, in the experiment involving aqueous 0.5 M Na₂SO₄ electrolyte system, it has been found that the capacitance for MNO - 3 sample is about 34 F/g and the energy density is about 23 Wh/kg for MNO - 3 sample. Finally, in the presence of aqueous-based 1 M Mg(ClO₄)₂ electrolyte, MNO - 3 sample is found to exhibit a capacitance of about 26 F/g and an energy density of about 17 Wh/kg, respectively. In all three electrolyte systems, the MNO -3 sample exhibit a long cycle capacitance retention of greater than 85% for 1000 charge-discharge cycles.     

Resolution and isolation of enantiomers of (±)‐isoxsuprine using thin silica gel layers impregnated with l‐glutamic acid,comparison of separation of its diastereomers prepared with chiral derivatizing reagents having l‐amino acids as chiral auxiliaries

Thin silica gel layers impregnated with optically pure l ‐glutamic acid were used for direct resolution of enantiomers of (±)‐isoxsuprine in their native form. Three chiral derivatizing reagents, based on DFDNB moiety, were synthesized having l ‐alanine, l ‐valine and S‐benzyl‐l ‐cysteine as chiral auxiliaries. These were used to prepare diastereomers under microwave irradiation and conventional heating. The diastereomers were separated by reversed‐phase high‐performance liquid chromatography on a C₁₈ column with detection at 340 nm using gradient elution with mobile phase containing aqueous trifluoroacetic acid and acetonitrile in different compositions and by thin‐layer chromatography (TLC) on reversed phase (RP) C₁₈ plates. Diastereomers prepared with enantiomerically pure (+)‐isoxsuprine were used as standards for the determination of the elution order of diastereomers of (±)‐isoxsuprine. The elution order in the experimental study of RP‐TLC and RP‐HPLC supported the developed optimized structures of diastereomers based on density functional theory. The limit of detection was 0.1–0.09 µg/mL in TLC while it was in the range of 22–23 pg/mL in HPLC and 11–13 ng/mL in RP‐TLC for each enantiomer. The conditions of derivatization and chromatographic separation were optimized. The method was validated for accuracy, precision, limit of detection and limit of quantification. Copyright © 2014 John Wiley & Sons, Ltd.     

1,2‐Dimethoxyethane Degradation Thermodynamics in Li−O2 Redox Environments

The reaction thermodynamics of the 1,2‐dimethoxyethane (DME), a model solvent molecule commonly used in electrolytes for Li?O₂ rechargeable batteries, has been studied by first‐principles methods to predict its degradation processes in highly oxidizing environments. In particular, the reactivity of DME towards the superoxide anion O₂^? in oxygen‐poor or oxygen‐rich environments is studied by density functional calculations. Solvation effects are considered by employing a self‐consistent reaction field in a continuum solvation model. The degradation of DME occurs through competitive thermodynamically driven reaction paths that end with the formation of partially oxidized final products such as formaldehyde and methoxyethene in oxygen‐poor environments and methyl oxalate, methyl formate, 1‐formate methyl acetate, methoxy ethanoic methanoic anhydride, and ethylene glycol diformate in oxygen‐rich environments. This chemical reactivity indirectly behaves as an electroactive parasitic process and therefore wastes part of the charge exchanged in Li?O₂ cells upon discharge. This study is the first complete rationale to be reported about the degradation chemistry of DME due to direct interaction with O₂^?/O₂ molecules. These findings pave the way for a rational development of new solvent molecules for Li?O₂ electrolytes.     

The Szegö kernel on a class of non-compact CR manifolds of high codimension

We generalize Nagel’s formula for the Szegö kernel and use it to compute the Szegö kernel on a class of non-compact CR manifolds whose tangent space decomposes into one complex direction and several totally real directions. We also discuss the control metric on these manifolds and relate it to the size of the Szegö kernel.     

Vitamin K1 distribution following intravenous vitamin K1–fat emulsion administration in rats

This study investigated vitamin K₁ (VK₁) distribution following intravenous vitamin K₁–fat emulsion (VK₁–FE) administration and compared it with that after VK₁ injection. Rats were intravenously injected with VK₁–FE or VK₁. The organ and tissue VK₁ concentrations were determined using high‐performance liquid chromatography method at 0.5, 2 and 4 h to determine distribution, equilibrium and elimination phases, respectively. In the VK₁–FE group, the plasma, heart and spleen VK₁ concentrations decreased over time. However, other organs like liver, lung, kidney, muscle and testis, reached peak VK₁ concentrations at 2 h. In the VK₁ injection group, the liver VK₁ concentrations were significantly higher than those in other organs at the three time points. However, VK₁ concentrations in the other organs peaked at 2 h. In addition, in VK₁–FE group, the heart, spleen and lung VK₁ concentrations were significantly higher than those in the VK₁ injection group at the three time points, and the liver VK₁ concentration was significantly higher than that in the VK₁ injection group at 4 h. The VK₁ amount was greatest in the liver compared with the other organs. Thus, the liver is the primary organ for VK₁ distribution. The distribution of VK₁ is more rapid when injected as VK₁–FE than as VK₁. Copyright © 2015 John Wiley & Sons, Ltd.