梯度法

词组短语

gradient method

与梯度法相关的网络例句 [注：此内容来源于网络，仅供参考]

Combined CD method and a new conjugate gradient method given by Liu.Y, Storey. C, the second class of nonlinear conjugate gradient method is proposed, which not only has descent property, but also is proved global convergence with the general Wolfe line search. Finally, the numerical results show this class of conjugate gradient methods is very efficient.

第二类算法是结合CD法和Liu.Y， Storey.C提出的新共轭梯度法，提出一类新的非线性共轭梯度法，新方法不但具有下降性质，而且在推广的Wolfe线搜索下是全局收敛的，最后进行了数值验证。

Meanwhile the biconjugate gradient method instead of the conjugate gradient method is used to accelerate the iteration process.

同时用双共轭梯度法代替共轭梯度法来加速迭代过程。

Among used machine learning methods, the gradient descent method is widely used to train various classifiers, such as Back-propagation neural network and linear text classifier. However, the gradient descent method is easily trapped into a local minimum and slowly converges. Thus, this study presents a gradient forecasting search method based on prediction methods to enhance the performance of the gradient descent method in order to develop a more efficient and precise machine learning method for Web mining.However, a prediction method with few sample data items and precise forecasting ability is a key issue to the gradient forecasting search method. Applying statistic-based prediction methods to implement GFSM is unsuitable because they require a large number of data items to model a prediction model. In the contrast with statistic-based prediction methods, GM(1,1) grey prediction model does not need a large number of data items to build a prediction model, and it has low computational load. However, the original GM(1,1) grey prediction model uses a mathematical hypothesis and approximation to transform a continuous differential equation into a discrete difference equation in order to model a forecasting model.

其中梯度法是一个最常被使用来实现机器学习的方法之一，然而梯度法具有学习速度慢以及容易陷入局部最佳解的缺点，因此，本研究提出一个梯度预测搜寻法则(gradient forecasting search method， GFSM)来改善传统梯度法的缺点，用来提升一些以梯度学习法则为基础的分类器在资讯探勘上的效率与正确性；而一个所需资料量少、计算复杂度低且精确的预测模型是梯度预测搜寻法能否有效进行最佳解搜寻之关键因素，传统统计为基础之预测方法的缺点是需要较大量的数据进行预测，因此计算复杂度高，灰色预测模型具有建模资料少且计算复杂度低等优点，然而灰色预测理论以连续之微分方程式为基础，并且透过一些数学上的假设与近似，将连续之微分方程式转换成离散之差分方程式来对离散型资料进行建模及预测，这样的作法不尽合理，且缺乏数学理论上的完备性，因为在转换过程中已经造成建模上的误差，且建模过程仅考虑相邻的两个资料点关系，无法正确反应数列未来的变化趋势。

The choice of step-length strongly affects the convergence rate of the Gradient Method, the classical Gradient Method—the method of steepest descent converges rather slowly in most cases, the poor behavior of the method is due to the optimal choice of step-length.

步长的选取对梯度法的收敛速度影响非常大，经典的梯度法-最速下降法在大多数情况下收敛得相当慢的原因在于最优步长的选取。

Analysis of the feature and geometry of the normal compaction trend is used to establish the compaction trend with polynomial of degree 2 or 3 or polygonal line, and to predict the distribution of geopressure using the method of seismic transit time deviation -pore pressure gradient in Yinggehai bas.

研究了压力预测的方法，推导地震声波时差差值—压力梯度法的计算公式并指出它的优点是能够充分利用已知的钻井实测压力资料；从理论上分析了正常压实趋势线的性质和形态，并用2~3次多项式和分段函数建立正常压实趋势线；根据莺歌海盆地高温高压及沉积速率高、泥底辟发育的特征，建立合理的压实趋势线，并用声波时差差值—压力梯度法预测了莺歌海盆地的压力分布。

The projection gradient method will be a possible way to solve the problem that we just get. It has been shown that the projections of the every directions, of which is the boundary point in linear restraint problems, are the possible decent directions, and the projection of negative grads direction is a decent direction. In 1960, Rosen proposed the basic idea of projection gradient methods, and then lots of researchers have been tried to find the convergence of this method. But most of them get the convergence with the condition to amend the convergence itself.

在约束最优化问题的算法中怎样寻找有效的下降方向是构造算法的重要内容，在寻找下降方向方面可行方向法中的投影梯度法有效的解决了下降方向的寻找问题，利用线性约束问题边界点的任意方向在边界上的投影都是可行方向，而负梯度方向的投影就是一个下降方向。60年代初Rosen提出投影梯度法的基本思想，自从Rosen提出该方法以后，对它的收敛性问题不少人进行了研究，但一般都是对算法作出某些修正后才能证明其收敛的，直到最近对Rosen算法本身的收敛性的证明才予以解决。

Until recently, the convergence of the Rosen"s arithmetic has been proved. The projection gradient method is the generalization of Steepest decent method to constraint problem. So it does have a faster convergence speed that the fastest decent method . To solve the problem, many researchers have generalizat the well-developed optimization with unconstraint to the Rosen"s gradient method. The conjugate gradient method is one of success to solve the problem.

投影梯度法是最速下降法对约束问题的推广，因此没有较快的收敛速度，为了解决这个问题很多中外学者把发展得比较成熟的无约束最优化算法作类似的推广，其中共轭梯度方法是近年发展的很成熟的方法，它具有计算简单，算法结构好，计算量少，具有良好的收敛性等优点，而Rosen投影梯度法的提出使寻找下降方向变得简单。

In the optimal control of kinematics for redundant robots, two new concepts, the matrix weightability measure and the self-motion declinability measure are proposed. Further, a modified weighted gradient projection method which combines gradient projection method with weighted least-norm solution is presented, it can attain ideal optimal effects at low velocities.

本文对冗余自由度机器人的运动学、动力学和容错等方面的优化控制进行了深入的分析和探讨：在冗余度机器人运动学优化控制方面，提出矩阵可加权度和自运动可衰度的概念，将梯度投影法和加权最小范数解法有机的相结合，形成可优化能力强、关节速度较低的改进的加权梯度法。

Addressing to deficiencies of this method,optimum partition method which is widely used in mass analysis is developed to fit thermocline boundary calculation.

针对垂直梯度法的不足，引入水团分析中的最优分割法，对典型剖面，以及边界型、逆变型、多层型等几类特殊剖面的跃层边界分别进行了计算，并对垂直梯度法与垂直梯度法的计算结果进行了比较分析，结果表明最优分割法确定的跃层边界更为合理。

Chapter 8: We discuss the acceleration of multigrid by Krylov subspace approaches, and recognize the reason for slow convergence of algebraic multigrid methods is that error reduction is significantly less efficient for some very specific error components which may cause a few eigenvalues of the algebraic multigrid iteration matrix to be considerably closer to 1 than all the rest. However, the eigenvectors belonging to the few isolated eigenvalues can be expected to be typically captured after only a few conjugate gradient iterations, which accelerate algebraic multigrid algorithms. Theoretical analysis and numerical results of some practical problems show the iterant recombination accelerates algebraic multigrid convergence.

第八章：先介绍用Krylov子空间迭代法加速一般多重网格方法收敛的基本框架，然后紧紧抓住引起代数多重网格方法收敛减慢的根本原因往往是误差减小对几个特别的误差分量不明显，这导致代数多重网格方法的迭代矩阵的几个特征值接近于1，而共轭梯度方法则能比较典型地捕捉属于孤立特征值的特征向量，从而推导出有效的共轭梯度加速算法和程序实现，不仅从一些具体的实际应用例子的数值结果去验证迭代复合加速收敛的效果，而且还从理论上分析了预处理共轭梯度法加速代数多重网格法收敛的机理。

与梯度法相关的网络解释 [注：此内容来源于网络，仅供参考]

conjugate gradient method：共轭梯度法

共轭梯度法(Conjugate Gradient Method)是以共轭方向(Conjugate Direction)作为搜索方向的一类算法. 最初的共轭梯度法由Hesteness和Stiefel于1952年为求解线性方程组而提出,后来用于求解无约束最优化问题.

conjugate gradient method：共扼梯度法

共轭梯度法:Conjugate gradient method | 共扼梯度法:Conjugate gradient method | 共轭梯度算法:Conjugate Gradient Algorithms

three-term conjugate gradient method：三项共轭梯度法

预条件共扼梯度法:Preconditioned Conjugate Gradient Method | 三项共轭梯度法:three-term conjugate gradient method | 非线性共轭梯度法:nonlinear conjugate gradients method

Preconditioned Conjugate Gradient Method：预条件共扼梯度法

多项式预处理共轭梯度法:Multinomial Preprocessing Conjugate Gradient Method | 预条件共扼梯度法:Preconditioned Conjugate Gradient Method | 三项共轭梯度法:three-term conjugate gradient method

Preconditioned Conjugate Gradient Method：预条件共轭梯度法

共扼梯度最优化:Conjugate Gradient Optimization | 预条件共轭梯度法:preconditioned conjugate gradient method | 多搜索方向共扼梯度方法:multiple search directions conjugate gradient

nonmonotone conjugate gradient method：非单调共轭梯度法

条件预优共轭梯度法:Pre- conditioned conjugate gradient method | 非单调共轭梯度法:nonmonotone conjugate gradient method | 三项共轭梯度算法:three-term conjugate gradient method

gradient method：梯度法arX中国学习动力网

gradient 倾斜度arX中国学习动力网 | gradient method 梯度法arX中国学习动力网 | gradient of temperature 温度梯度arX中国学习动力网

general reduced gradient method, GRG method：通用简约梯度法

gradient method 梯度法 | general reduced gradient method, GRG method 通用简约梯度法 | feasible path method 可行路径法

biconjugate gradient method：双共轭梯度法

共轭梯度方法:conjugate gradient method | 双共轭梯度法:biconjugate gradient method | 温度梯度法:thermal gradient method

SSOR preconditioned CG method：对称超松弛预处理共轭梯度法

梯度法:Gradient-based method | 对称超松弛预处理共轭梯度法:SSOR preconditioned CG method | 三参数共轭梯度法簇:A family of three parameter conjugate gradient medhod