1) Local Kernel
局部核函数
1.
To improve the performance of QSAR model,a novel non-lineal combinatorial method named GK-LK-SVR based on global kernel and local kernel was proposed in this paper.
为提高定量构效关系(quantitative structure-activity relationship,QSAR)模型预测的精度,以支持向量回归(support vector regression,SVR)全局与局部核函数,发展出1种非线性组合方法GK-LK-SVR,其基本思路为:依均方误差(MSE)最小原则,分别基于SVR的全局与局部核函数筛选描述符后预测,实测值与不同核函数的预测值组合成混合样本,然后再依MSE最小原则基于SVR对混合样本实施核函数寻优及子模型筛选,最后以留一法完成预测。
2) local kernels
局部性核函数
3) nonlocal kernel function
非局部核函数
1.
Based on the nonlocal theory the correlation is studied between the nonlocal kernel function and the simulating result for the strain-localization behavior of brittle materials.
在非局部理论框架下,研究了应变局部化现象模拟结果与所用非局部核函数的相关性。
4) locally Lipschitz function
局部Lipschitz函数
1.
In this paper,the solution existence for quasilinear hemivariational inequality was analyzed using the variational method and the nonsmooth critical point theory of the locally Lipschitz function.
我们的方法是变分法及局部Lipschitz函数的非光滑临界点理论。
2.
This paper discusses the generalization of the deformation theorem and its application,and some new critical point theorems of locally Lipschitz functions are given based on some improved classical critical point theorems.
证明了一个形变定理,并由此得到局部Lipschitz函数的几个临界点定理,其结果改进了几个经典的临界点结论。
3.
In the present paper,some minimax theorems of locally Lipschitz functions are given by the Ekeland variational principle and tow critical point theorems are improved.
文章由Ekeland变分原理得到局部Lipschitz函数的几个极大极小定理,并改进了已有的两个临界点定理。
5) local base function
局部基函数
6) local bubble function
局部bubble-函数
1.
This paper is devoted to the development of stabilized finite element methods by empolying local bubble functions for adveetive-diffusive models which has the form σu+a·(?)u-k△u =f.
本文针对形如σu+α·u-kΔu=f对流—扩散型的模型问题,发展耦合局部bubble-函数的有限元方法,我们就α=0和σ=0两种情形证明了方法的与“影响因素”σ和pedlet-数无关稳定性及全局最佳收敛阶。
补充资料：函数的局部逼近
函数的局部逼近
local approximation of fimctions
函数的局部逼近【】以川a即rO:应na石阅of加叫出创旧;二oK幼‘。oe nPo6二H二eu，e中yllK颐，益」 集合EC=R“上函数f的一种逼近度量(特别是最佳逼近(比tapproximation)度量).主要问题是研究当m巴E~O时一个函数局部逼近的性态.在某些情形下，可借助函数的局部逼近来刻画被逼近函数的光滑阶，设E。(f;(:，刀))为区间(:，刀)(a蕊:<刀(b)上。次代数多项式对函数fcC【a，b]的最佳逼近.下述结论成立:函数f在la，b]上各点有。十1阶连续导数的充分必要条件是 奥琴兰典真卫一月‘x、，a簇x簇“· 气P一“夕对口~x，，一x，:
说明：补充资料仅用于学习参考，请勿用于其它任何用途。
参考词条