In this paper an interval iterative method is given for seeking the global maximum of a second order differentiable function.
本文就二次可微函数类给出一类求大范围极值的区间迭代方法。
By very elementary method, it is proved that every continuous differentiable function defined in a perfect set has differentiable extension.
本文完全用初等的方法证明了完全集上连续可微函数都有可微开拓。
It is also proved as application that any bounded variation function can be seen as a differentiable function beside a set of arbitrary small measure.
作为应用，证明了任意有界变差函数都与某可微函数在除过测度任意小的集合外重合。
Every Layer has a simple API: it transforms an input 3d volume to an output 3d volume with some differentiable function that may or may not have parameters.
每一层都有一个简单的API：用一些含或者不含参数的可导的函数，将输入的三维数据变换为三维的输出数据。
A type of continuous and Non-differentiable Function is studied, its characteristics and essence are revealed, and its condition of construction is presented.
对一类连续但不可导函数进行研究，揭示其特点和本质，并给出其构造条件。
For functions of them, the delta method implies the following result, proved in Section 14.1.4: Let denote a differentiable function of, with sample value for a multinomial sample.
为作用的他们，三角洲方法暗示以下结果，被证明在部分14.1.4：让表示一个能区分的作用，以样品价值为一个多项式样品。
Some Suggestions for executing this method are presented, graphical results of nowhere differentiable continuous function are shown.
对此，阐明了一些有效的建议，并给出了无处可微连续函数的图示结果。
Rational number can approximate to real number, use the notation of approximate one can prove Riemann function isn t differentiable anywhere, that the Rational points are dense in unit circle.
利用有理数对实数逼近的表示方式，给出黎曼函数处处不可导的一种证明，给出单位圆周上的有理点在单位圆上稠密的证明。
Rational number can approximate to real number use the notation of approximate one can prove riemann function isn't differentiable anywhere that the rational points are dense in unit circle .
摘要利用有理数对实数逼近的表示方式，给出黎曼函数处处不可导的一种证明，给出单位圆周上的有理点在单位圆上稠密的证明。
This article was to offer necessary and sufficient condition and application of conjugate differentiable complex function in complex system.
给出了复形式的复变函数的共轭可微性充要条件的证明及其重要应用。
The traditional bi-level model formulation of these problems is transformed into a single-level continuously differentiable problem using a marginal function approach.
并提出了一种基于路段变量的连续可导的边际函数和距离函数，从而把两层规划转化成单层规划问题。
Because the optimization problem may be discontinuous and non-differentiable even has no objective function, the traditional optimization methods are unable to tackle with it.
由于优化问题可能是不连续的、不可微的甚或是没有函数解析式的，传统经典的无约束优化方法在应用时会受到限制。
Because the error transfer function of rough neural network is not differentiable, genetic algorithms are applied for training the network.
由于粗神经网络的误差传递函数不可微，所以采用遗传算法来训练粗神经网络。
Minimax problem is a sort of non-differentiable optimization problem and the entropy function method provides a efficient approach to solve such kind of problems.
极大极小问题是一类不可微优化问题，熵函数法是求解这类问题的一种有效算法。
The value function turns out to be continuous in the whole plane, and is even continuously differentiable in the interiors of both right and left half planes.
说明了值函数在整个平面上是连续的，在左右两个半平面的内部还是连续可微的。
For the learning process, a new kind of empirical risk function is proposed which is differentiable and can be minimized by gradient descent strategy.
对于参数的学习，提出了一种适用于分类器的可微经验风险函数，该函数能够有效地利用梯度下降法进行最小化。
For the learning process, a new kind of empirical risk function is proposed which is differentiable and can be minimized by gradient descent strategy.
对于参数的学习，提出了一种适用于分类器的可微经验风险函数，该函数能够有效地利用梯度下降法进行最小化。
应用推荐