查询词典 uniformly convergent

The main results of this research are as follows:(1) A parallel algorithm for solving dense matrix linear equations on distributed-memory multi-computer is presented. Convergent velocity of this algorithm is better than parallel Jacobi method"s, and parallelism is better than Gauss-Seidel method"s. According to theoretical analysis, convergent conditions are given. Moreover numerical experiments indicate that results agree with theory.

主要完成了如下研究工作：(1)提出了一种求解系数矩阵为稠密矩阵的线性方程组的适合于MIMD分布式存储的并行迭代算法，该算法设计上兼取了Jacobi算法并行性好的优点和Gauss-Seidel算法收敛速度快的长处，并且在理论上给出了该算法成立的收敛性条件，并通过数值计算验证了此算法结果与理论相符合。

From the matrix theory and the general principle of the iterative convergence of the linear coupled equations, this dissertation firstly proved the sufficient necessary convergent conditions for CMAC batch learning as well as the necessary convergent conditions for CMAC incremental learning, in which no additive conditions were needed. The past researchers'conclusions under the special condition of the positive articulation matrix are generalized and improved.

从矩阵理论和线性方程组迭代收敛的一般性原理出发，首次在不附加特殊条件的情况下，证明了CMAC学习算法在批量和增量两种学习方式下收敛的一般性定理，给出CMAC算法在批量学习方式下收敛的充要条件，以及在增量学习方式下收敛的必要条件，对前人在关联矩阵正定的特殊条件下得出的结论进行了推广和改进。

Finally,the inequality of maximal operators of tree martingale transformsis applied for obtaining the UMD space valued Vilenkin system and the sum ofVilenkin-Fourier series of UMD space valued Vilenkin system is convergent a.e.or convergent in LP-norm.

最后，应用UMD空间值树鞅鞅变换的极大算子不等式，我们获得了UMD空间值Vilenkin系统以范数收敛和几乎处处收敛定理，即Carleson定理的扩展。

The proposed algorithm is convergent to the solution through the solutions of a series of condensation trust region subproblems. It is proved that the method is globally convergent and that the convergence rate is quadratic.

算法通过求解序列压缩信赖域子问题来获得原问题的最优解，并具有全局收敛性和局部二阶收敛速度。

The necessary and sufficient conditions for convergent of characteristic domain and the conditions for tangent plane continuity are obtained. The conditions for having limited value of the second order of differential at the convergent point are shown as well.

给出了任一特征域收敛的充要条件；给出了曲面在任一极限收敛点上达到一阶几何连续的条件；分析了极限收敛点处的二阶连续，给出了二阶导数有界的条件。

On the basis of the advantages of the previous distillation calculation algorithms, by eliminating the non—zero elements to the two sides of the tridiagonal band of elements in the component material balance equations by correlation of the draw rate from the main column to the sidestrippers and the return vapor stream from the sidestrippers to the main column, it changes these equations into a typical form of the tridiagonal matrix, and uses the simultaneous convergent strategy of the main column and the side strippers to speed the algorithm convergent rate.

在吸收前人蒸馏计算算法优点的基础上，通过求解主塔抽出到汽提塔的量和汽提塔返回主塔的蒸汽流之间的关联，以便消除组分物料平衡方程组系数阵中的三对角线两边的非零元素，使它转化为典型的三对角矩阵形式，而且使用主塔和汽提塔同时收敛策略，加快了算法收敛速度。

The effective stimulating sites were indentified by defensive cardiovascular response and other components of defence reaction evoked by 3 s trains of rectangular cathodal pulses (100Hz, 0.5ms pulse duration, 70 -150μA intensity for dPAG and 150-250μA for PF) delivered from a programmed multichannel constant current stimulator. The responses included an increase in blood pressure and heart rate, pupillodilation and an increase in the rate and depth of respiration. Currents of 1Hz, intensity of 100- 150μA for dPAG and 200-250μA for PF were used to observe the effect of the stimulation of them on RVLM neurones. The deep peroneal nerve and superfacial deep peroneal nerve were carefully separated, and put on a pair of stimulating electrode respectively, Both nerves were stimulated with constant rectangular pulses of 1Hz, 0.5ms duration and 300μA current intensity to see their effects on RVLM neurones. The baro-and chemo-receptors were activated by the intravenous injection of phenylephrine or 0.208 M NaH2PO4. The degree of convergent inputs to individual RVLM neurone was studied by the stimulation of the sites mentioned above. It was found that neurones about 61.4%(27/44) and 35.3%(6/17) received convergent inputs from three and four places respectively.

过去的研究主要集中于整体水平，利用电刺激或化学刺激RVLM神经元，观察对血压、心率及交感活动的影响，并对影响心血管交感活动的神经递质进行了广泛探讨，积累了大量资料；在神经元水平，主要研究刺激脑内核团、外周神经、化学感受器及压力感受器等对RVLM神经元放电的影响和某些神经递质影响神经元活动的机制、然而，对RVLM神经元如何整合不同信息、其可能突触机制等研究得较少，本互作拟采用在体细胞内、细胞外记录电活动与微电泳，结合中枢核团及外周神经电刺激等方法，对RVLM神经元在不同状态下的整合作用规律与可能机制进行初步探讨，为进一步了解这一复杂的整合过程提供新的理论依据。

Some of convergent theorems and convergent comparison theorems areproved,and some of examples illustrate that these kinds of preconditioned methods iseffective.

文中具体的数值例子可说明本章的两种预处理方法是有意义的。

Aiming at the slow convergent rate of standard Q-learning, multi-step on-policy SARSA reinforcement algorithm is adopted. This algorithm has some characters of faster convergent rate, smooth running track and safer training process than standard Q-learning.

针对标准Q-学习收敛速度较慢的缺点，采用多步在策略SARSA强化学习算法，该算法克服了Q-学习收敛速度慢的缺点，运行轨迹平滑，训练过程相对比较安全。

Simulation result shows that X-ray spot size is almost certainly convergent when the number of beamlets increases gradually,and the convergent size is about 1.03 mm.

计算结果表明：子束流层数一定时，焦斑直径波动在0.03mm范围内；随子束流划分层数的不同，计算所得焦斑直径的最大不确定度为±0.05mm，而焦距几乎不变化，其波动在0.08cm范围内；随子束流划分层数的增加，焦斑直径计算结果是收敛的，最终收敛值约为1.03mm。

The production moved off-Broadway to New York City, where she won the Joseph E. Calloway award for her performance, and was profiled in American Theater Magazine as one of ten young theater artists to watch.

生产提出过百老汇的新的纽约市，在那里她赢得了约瑟夫calloway奖，她的表现，并于异型在美国的战区杂志之一， 10战区的青年艺术家观赏。

It's a risk tradeoff, and I believe that adding test support code usually wins.

这是一个风险的权衡，我相信添加测试代码通常会占上风。