查询词典 Hessian

The condition number of modified Lagrange function's Hessian is estimated, which also depends on the penalty parameter.

最后估计了修正Lagrange函数的Hesse阵的条件数，它同样依赖于罚参数。

In this paper, we use Constant Rank Theorem to give some convexity results for a class of ...

本文中，我们将利用常秩定理对一类Hessian方程给出几个凸性的结果。

Investigate the numerical solution of the Dirichlet Problem of functions of the eigenvalues of the Hessian with the theory of viscosity solution and the solution is continuous.

第五章研究了将网络参数的混合算法和网络规模的调整算法两个算法有机结合，构建求解微分方程定解问题的理想RBF网络。

The dissertation focuses on performance gradient estimation and SRO algorithm of DES aswell as other related topics,such as modeling of DES and discrete event simulation,performancemeasure Hessian matrix estimation,Stochastic Approximationand so on.General Semi-Markov Processis adapted to describe behaviour of DES and discrete eventsimulation.A general result is obtained and applied to analyze strong consistency of performancegradient Perturbation Analysisestimator.Conditions for effectiveness of InfinitesimalPerturbation Analysisare also discussed.New PA algorithms,such as server PA,PA usingDirac generalized function,job sequence PA etc.,are proposed to deal with discreteparameters.Furthermore,strategies of combining PA algorithms with certain iterative rules areanalyzed widely in order to optimize DES parameters and some SRO algorithms are constructed toslove DES optimization problems.

论文将DES性能分析领域中性能梯度估计和单运行仿真优化作为主要的研究对象，探讨了与性能梯度估计有关的DES和离散事件仿真过程建模、高阶导数估计以及随机逼近等理论和方法；建立了具有一般性的DES和离散事件仿真的广义半Markov过程模型；获得了分析性能梯度估计量强相合性的理论结果和无穷小摄动分析算法应用的条件；提出了诸如服务台数目的摄动分析、奇异摄动分析、串联生产线和加工车间中工件加工序的摄动分析等一系列摄动的思想和算法，对连续参数和离散参数都进行了深入的研究；对摄动分析算法同优化规则结合实现仿真优化的策略也进行了广泛的分析，构造了新型的变界截尾的随机逼近算法，并用于DES的单运行仿真优化研究；最后，给出了DES性能分析方法的实证性研究结果。

And introducing Levenberg-Marquardt method to perturb the Hessian matrix is also quite unique and successful.

大量的实验结果也证实了我们两种方法的有效性。

And the Levenberg-Marquardt iteration, which can avoid to compute Hessian matrix, is adopted as the learning rule to train the samples of this feedforward network.

针对传统的梯度下降优化算法收敛速度慢的特点，采用可避免计算Hesse矩阵的Levenberg-Marquardt算法来训练神经网络。

In order to perform an accurate navigation, the centerline of the model must be extracted. In this paper, a new centerline extraction algorithm based on Hessian Matrix is proposed.

要想快速准确地进行虚拟内窥漫游，一个首要的步骤是要先提取出模型的中心线以指导视点的移动。

Compared to perturbation method, the influences of many factors on voltage stability margin are considered and Hessian matrix of control parameters is not necessarily calculated. The method has high calculation speed and accuracy.

相对于摄动方法而言，该方法将各种因素对于电压稳定域的影响考虑在内，且不需要求取控制参数的海森矩阵，具有计算速度快、精度高的特点。

The Hessian matrix of every function in this model is constant, so it will be calculated once in the entire optimal process based on interior point method, which speeds up each iteration.

该新模型的海森矩阵是精确的常系数矩阵，在内点法迭代过程中只需要计算一次，从而缩短了每次迭代的计算时间。

In Chapter 5, a new method with direct partitioned column updates of cholesky factorization for solving large scale sparse unconstrained optimization problems is proposed, which depends on a consistent partition of the columns and cholesky factorization of the Hessian matrix.

第5章基于稀疏Hesse阵的列的相容分划，并在其Cholesky分解的基础上建立了求解大型稀疏无约束优化问题的直接校正Cholesky因子算法。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。