核函数

It was based on t he Gauss distribution and conformal transformation method presented by Amari and Wu.

分析了确定数据分布特征的重要性，给出了判断数据呈高斯分布的方法，探讨了SVM核函数及其参数选择与数据分布的相关性。

Based on his result, Amari proposed a method of modifying a kernel function using it抯 geometric character. Which can improve the respective of SVM.

根据他的结果，Amari在同年提出了一种利用核函数的几何性质对其进行修正的方法，以提高SVM的分辨率。

Hence, reproducing kernel function has not only the advantage of calculation, e.g.it can transform the integration into the sum of finite function values, but also theadvantage of local characteristic of wavelet function.

所以再生核函数既具有计算优势，如能将积分计算准确地转换成有限个函数值的和来计算，又具有小波函数的局部刻画优势。

This method uses the kernel function to map the nonlinear space into a linear high dimension space.

该方法使用核函数完成非线性空间到高维线性空间的映射，避免了高维空间中的数据处理和非线性映射函数的使用。

In this paper we present a local self-concordant barrier function based on a kernel function in the neighborhood of the central path of linear optimization problem.

在这篇文章中，我们介绍了基于核函数的局部self-concordant障碍函数，它在线性优化问题的中心路径及其邻域内满足self-concordant性质。

Secondly,we consider the binary classification problem by learning from samples drawn from a non-identical sequence of probability measures.

在核函数为欧氏空间上的高阶光滑函数时，e~1正则化算法的学习率还可以进一步改进。

One is for Green's function in elements by using linear integrals; the other is for its normal derivative by the divergence theorem.

作法是仅需对原有的奇异性核函数加入一项使其异性消除，同时再将函数的解析积分值减去原方程式。

The experimental result shows that in the spectra region between 6500 cm~(-1) and 5300 cm~(-1),the best model is built by the standard normal variate ...

试验结果显示，通过标准归一化预处理，选取6500~5500 cm-1波长范围内的光谱经过主成分分析后，提取11个主成分，选用径向基函数作为核函数建立的模型最佳。

The main works include five parts:(1) A modified K-means algorithm for optimizing the RBPNNs' structures was proposed;(2) The RBPNNs' structure optimization method based on the orthogonal least square algorithm was proposed that can greatly reduce the computation expense for the structure optimization;(3) Using the GA to perform the whole structure optimization of the RBPNNs was proposed in this thesis, which include simultaneous optimization of the hidden centers and the corresponding controlling parameters of the kernel functions.

主要工作体现在如下三个方面：(1)提出一种改进的 K-Means 算法；(2)提出了基于递推正交最小二乘算法的结构优化算法来训练径向基概率神经网络，从而大大减小了用于结构优化的计算开销；(3)提出使用遗传算法来实现径向基概率神经网络的全结构优化，即隐中心矢量和核函数控制参数同时优化，通过新设计的编码方式、新构造的适应度函数，充分发挥了 GA 的全局搜索性能，使得所优化的径向基概率神经网络的结构趋于最简。

A generic face mesh was deformed and the texture image was computed using image-based modeling technique. The deformed mesh was then convolved with a kernel function to generate a convolution surface, and wrinkles were generated by modulating the surface with a designed profile function.

运用基于图像的建模方法将通用人脸网格变形并计算出纹理图像，变形后的网格再与一个核函数卷积生成卷积曲面，通过轮廓函数调整曲面的形状而产生皱纹。

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。