1) Reduced Projected Hessian
![点击朗读](/dictall/images/read.gif)
既约Hesse阵
2) Hessian
[英]['heʃən] [美]['hɛʃən]
![点击朗读](/dictall/images/read.gif)
Hesse阵
1.
Recently, it is a hot research at home and abroad in portioning groups of the matrix to iteration which is based on the structure of the Hessian.
近年来,结合稀疏Hesse阵的结构,按照分划组的思想将矩阵列分划进行迭代是国内外研究的一个热点。
2.
In Chapter 4, a symmetric triangular partitioned secant method for solving large scale sparse unconstrained optimization problems is constructed, which is a combination of a secant method and a finite difference method, and depends on a consistent partition of the columns of the lower triangular part of the Hessian matrix; it reduces the number.
此算法是基于稀疏Hesse阵的下三角部分的列的相容分划的基础上建立起来的,把割线法和有限差分法有机的结合在了一起,在每次迭代中,把由Powell和Toint的算法(间接下三角替换法)所需要的梯度赋值次数减少了一。
3) Hesse matrix
![点击朗读](/dictall/images/read.gif)
Hesse矩阵
1.
Hesse Matrix of Sufficient Condition for Extrema of n-variable implicit function;
![点击朗读](/dictall/images/read.gif)
判定n元隐函数取极值的充分条件Hesse矩阵
2.
Hesse matrix of sufficient condition for conditional maximum and minimum;
![点击朗读](/dictall/images/read.gif)
判定条件极值的充分条件的Hesse矩阵
3.
This paper presents an approximation calculation for Hesse matrix to train networks.
![点击朗读](/dictall/images/read.gif)
本文讨论当误差不为零或者不为线性函数,即二阶项S(w)不能忽略时的Hesse矩阵的近似计算,进而训练网络。
4) irreducible morphism
![点击朗读](/dictall/images/read.gif)
既约映射
5) reduced road network
![点击朗读](/dictall/images/read.gif)
既约路网
6) reduced gradient
![点击朗读](/dictall/images/read.gif)
既约梯度
补充资料:既约多项式
又称“不可约多项式”。次数大于零的有理数系数多项式,不能分解为两个次数较低但都大于零的有理数系数多项式的乘积时,称为有理数范围内的“既约多项式”。在实数或复数范围内,也有相应的定义。实数范围内的既约多项式是一次或某些二次多项式,复数范围内的既约多项式必是一次多项式。
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