核函数

We apply the kernel methods of regression functions and conditional density functions to obtain a series of as the estimators of . Then using Monte Carlo Analysis method , we can get the optimal choice of the unknown parametric that is . Examples illustrate the models and will show this estimation is effective.

应用回归函数核估计和条件密度函数的核估计方法估计未知量，将得到的一组估计值用Monte Carlo Analysis方法处理便得到最优的，并通过算例模拟讨论以上估计方法的优劣。

In order to avoid the singularity in the integration, the integrand was expressed to be sum of complex exponential terms by using discrete complex image method. With generalized pencil of function method introduced, the number, location and intensity of images were obtained without extracting the quasi-static and surface wave terms in the integrand, which makes DCIM more efficient for multi-layered media.

为了避免积分中的奇异性，利用离散复镜像法将积分核用复镜像的指数求和式表示，引入广义函数束方法，可以在不提取积分核中表面波项的条件下，采用数值方法提取准静态项，给出复镜像点的数目、位置和强度，使得该方法在多层介质情况下对格林函数的计算更为有效。

In chapter III,we consider the wavelet method for the approximation of the nu-merical Bergman kernel.It is very clear that the Bergman kernel plays an importantrole in the complex analysis and elliptic equations.The Bergman kernel has a closedrelation with the Riemann mapping,Green function,Neumann kernel and the Szeg〓kernel etc.

第三章研究了数值Bergman核数值逼近的小波方法，Bergman核在复分析和椭圆型方程中占有重要地位，它与Riemann映射、Green函数、Neumann函数、核等有着密切联系。

By using the least square method and the weighted functionmethod of nonparametric estimation together, this paper defines the wavelet estimators of β and g.

本文采用了半参数估计常用的办法之一——最小二乘法结合非参数权函数估计法定义了参数分量和非参数分量的估计，这里的权函数不是通常所采用的核权函数或近邻权函数，而是由小波方法所得到的权函数，我们称它为小波权函数。

The new methodof defining spline function will generalize the concept to the more universal Hilbert space. Hence, we take into consideration the interpolation spline function, general linear differential operater in semi- open interval and the Green function defined accordingly, and the reproducing kernel to produce the Green function and prove the existence of the reproducing kernel.

采用了一种新的样条函数的定义方法，把样条函数的概念推广到更一般的Hilbert空间上，再考虑半开闭区间a，+∞上的插值样条函数，一般线性微分算子及其定义的Green函数和再生核，构造出Green函数并且证明了再生核的存在性。

This thesis will be expanded from two aspects: first, delta-sequence obtained by cubic spline interpolation cardinal function showed the properties of symmetry, Riesz basis and interpolation. Non linear convection diffusion equation (Burgers′equation) was used as an example. Then we modified this delta-sequence to improve the attenuation, the same example was used in numerical application.

在小波方法与PDE算法相结合方面，本文主要从两方面入手：首先是以样条插值基函数为基础构造δ-序列核，以非线性对流扩散方程为例给出了一种数值算法；进而对三次样条插值基函数进行改进，构造新的具有更快衰减性质的δ-序列核，同样以非线性对流扩散方程为例验证了新函数在数值求解中的有效性。

This chapter proposes three conceptions, i.e., Kernelled quasidiferential, star-kernel and star-diferential, and establishes their operational properties. A sufficient theorem and a sufficent and necessity theorem for a quasi-kernel being a kernelled quasidiferential are proven. Both the existence of star-kernel for a quasidiferentiable function and the existence of star-differential for a direnction-ally diferentiable function are established.

在这一章里，首先给出核拟微分，星核与星微分的定义及其它们的运算性质；然后证明了拟核微分的一个充分条件定理及一个充要条件定理；最后讨(来源：A27BC论文网www.abclunwen.com)论拟可微函数星核的存在性及方向可微函数星微分的存在性以及Penot-微分与上下导数之间的关系。

Thischapter proposes three concepts,i.e.,kernelled quasidifferential,star-kerneland star-differential,and establishes their operational properties.A sufficiencytheorem and a sufficiency and necessity theorem for a quasi-kernel being a ker-nelled quasidifferential are proven.Both the existence of star-kernel for a qua-sidifferentiable function and the existence of star-differential for a directionallydifferentiable function are established.The relationships between sub-and su-per derivatives and Penot differentials are discussed as well.

这一章我们引入了三个概念，即核拟微分，星核与星微分并建立了它们的运算性质；证明了拟核为核拟微分的一个充分条件定理及一个充分与必要条件定理；建立了拟可微函数星核的存在性及方向可微函数星微分的存在性并讨论了Penot微分与上、下导数的关系。

In NWFM,to solve self-consistently the set of DFT Kohn-Sham equations,nuclear weight function is employed for numerical solution of multicenter integra-tion,which decomposes the problem of three dimensional integration to a sum ofone-center,atomic-like integrations.

首先介绍如何利用线性组合的原子轨道方法求解密度泛函理论中的K-S方程；接着详细地说明了核权重函数的概念，并列出了两种常用的核权重函数形式，同时阐述了如何利用核权重函数求解包含多中心积分的矩阵元以及复杂的三维泊松方程以获得电子间库仑势的方法；然后给出了求解矩阵元、有效势以及体系总能的具体数值计算公式。

The number of support, vectors is decreased by 28%. Then, the structure of the Riemannian geometry is introduced in the input space, and using the Riemannian geometric modifies the kernel function of the classifier and gets less improved support vectors at the second training.

在第一次训练后产生的支持向量的基础上，将黎曼几何结构引入到输入空间，利用黎曼几何结构将分类器中的核函数进行修改，在第二次训练中再次减少了支持向量数目。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。