Non‐empirical calculations of NMR indirect carbon–carbon coupling constants. Part 4: Bicycloalkanes

A systematic study of the one‐bond and long‐range J(C,C), J(C,H) and J(H,H) in the series of nine bicycloalkanes was performed at the SOPPA level with special emphasis on the coupling transmission mechanisms at bridgeheads. Many unknown couplings were predicted with high reliability. Further refinement of SOPPA computational scheme adjusted for better performance was carried out using bicyclo[1.1.1]pentane as a benchmark to investigate the influence of geometry, basis set and electronic correlation. The calculations performed demonstrated that classical ab initio SOPPA applied with the locally dense Dunning's sets augmented with inner core s‐functions used for coupled carbons and Sauer's sets augmented with tight s‐functions used for coupled hydrogens performs perfectly well in reproducing experimental values of different types of coupling constants (the estimated reliability is ca 1–2 Hz) in relatively large organic molecules of up to 11 carbon atoms. Additive coupling increments were derived for J(C,C), J(C,H) and J(H,H) based on the calculated values of coupling constants within SOPPA in the model bicycloalkanes, in reasonably good agreement with the known values obtained earlier on pure empirical grounds. Most of the bridgehead couplings in all but one bicycloalkane appeared to be essentially additive within ca 2–3 Hz while bicyclo[1.1.1]pentane demonstrated dramatic non‐additivity of ?14.5 Hz for J(C,C), +16.6 Hz for J(H,H) and ?5.5 Hz for J(C,H), in line with previous findings. Non‐additivity effects in the latter compound established at the SOPPA level should be attributed to the through‐space non‐bonded interactions at bridgeheads due to the essential overlapping of the bridgehead rear lobes which provides an additional and effective non‐bonding coupling path for the bridgehead carbons and their protons in the bicyclopentane framework. Copyright © 2003 John Wiley & Sons, Ltd.     

神经网络在缺陷识别中的应用

为了准确、快速地进行缺陷识别，介绍了一种新型的前馈神经网络模型，即径向基概率神经网络。与以往的算法相比，该方法具有分类识别精度高且速度快的优点。仿真获得了很好的结果。     

The Gram Matrix of a PT-Symmetric Quantum System

The eigenstates of a diagonalizable PT-symmetric Hamiltonian satisfy unconventional completeness and orthonormality relations. These relations reflect the properties of a pair of bi-orthonormal bases associated with non-hermitean diagonalizable operators. In a similar vein, such a dual pair of bases is shown to possess, in the presence of PT symmetry, a Gram matrix of a particular structure: its inverse is obtained by simply swapping the signs of some its matrix elements.     

Prime divisors of sparse integers

We obtain a lower bound on the number of prime divisors of integers whose g-ary expansion contains a fixed number of nonzero digits. This revised version was published online in August 2006 with corrections to the Cover Date.     

多小波子空间上的单小波表示

本文在较弱的条件下,建立了2重多小波子空间与单小波子空间的关系.即由2重多小波构造出单小波.一方面,这种单小波的平移伸缩与2重多小波的平移伸缩生成的子空间是完全相同的;另一方面,它具有插值性.因此通过构造出的单小波建立了多小波子空间上的Shannon型采样定理.     

Banach空间上的框架与Riesz基

本文讨论Banach空间上框架、无冗框架与Riesz基之间的关系及它们的稳定性．     

反恐怖安全检测技术的物理基础

反恐怖安全检测技术已成为人们的热门话题。安全检测技术的诞生和发展离不开物理学家的贡献。物理学为反恐怖安全检测孕育了多种手段和方法。文章简要介绍了X射线探测、γ射线探测、四极共振分析、脉冲式快中子分析、离子迁移率光谱测量、毫米波探测、声学探测、磁场探测、电磁脉冲探测等安全检测技术的物理基础。     

The Lascoux, Leclerc and Thibon algorithm and Soergel's tilting algorithm

We generalize Soergel's tilting algorithm to singular weights and deduce from this the validity of the Lascoux-Leclerc-Thibon conjecture on the connection between the canonical basis of the basic submodule of the Fock module and the representation theory of the Hecke-algebras at root of unity. Supported in part by Programa Reticulados y Ecuaciones and by FONDECYT grant 1051024.     

Hilbert Space of Probability Density Functions Based on Aitchison Geometry

The set of probability functions is a convex subset of L1 and it does not have a linear space structure when using ordinary sum and multiplication by real constants. Moreover, difficulties arise when dealing with distances between densities. The crucial point is that usual distances are not invariant under relevant transformations of densities. To overcome these limitations, Aitchison＇s ideas on compositional data analysis are used, generalizing perturbation and power transformation, as well as the Aitchison inner product, to operations on probability density functions with support on a finite interval. With these operations at hand, it is shown that the set of bounded probability density functions on finite intervals is a pre-Hilbert space. A Hilbert space of densities, whose logarithm is square-integrable, is obtained as the natural completion of the pre-Hilbert space.     

Optimally Stable Multivariate Bases

We show that the tensor product B-spline basis and the triangular Bernstein basis are in some sense best conditioned among all nonnegative bases for the spaces of tensor product splines and multivariate polynomials, respectively. We also introduce some new condition numbers which are analogs of component-wise condition numbers for linear systems introduced by Skeel.