演算

This paper presents a maximum criterion-based phylogeny analysis method for unrooted trees. To increase the confidence level of the derived evolutionary trees, we use bootstrapping and consensus analysis in conjunction with a genetic algorithm for crossover and mutation between evolutionary trees.

本论文采用最大概率字元特徵式的评分方式，以及无根的建构方法，利用基因演算法对所有可能的树的子树做交配与突变，再使用统计自序法与一致演化树法来增加树的可信度。

For the infeasibility or unsolvability of the "Mixed Integer Programming" formulations, this study develops a heuristic algorithm to solve "Mater Planning" problem with fixed setup cost and time.

基於整数规划的不可行性，本研究提出一启发式演算法以解决主规划排程的问题，除了满足多最终产品的产品结构、复杂的供应链网路架构和多张订单需求、多期与共用料的环境，并加入整备成本与时间的考量规划整体供应链的生产与配送计画。

Hence, a self-balancing controller is implemented by considering the motion as well as configuration variation of the upper body (the 9-DOF mechanism). The control law utilizes the principle of computed torque method with online identification of related parameters using various sensors including an accelerometer. The incline angle is obtained by fusing a gyroscope and a tilt sensor. Under the balancing control, the forward motion of the robot is achieved by giving a desired tilt angle profile.

除此之外，本作品将此机器人装置在一具二轮式自动平衡移动平台之上，它提供了机器人灵活移动和转身的能力，但碍於此一平台本身为不稳定系统，所以必须实现一自动平衡演算法去计算底座以及上半身机器人动作行为的转动力矩，利用计算转矩法以及陀螺仪、倾斜仪和加速规去即时更新系统参数，让整个系统可以即时地保持平衡，并且利用设计角度值的概念让整个系统移动。

We find that the combination of the FFT method with the martingale control variate method is very useful to reduce the computational time while preserving the accuracy of simulations.

我们发现结合快速傅立叶演算法与马丁格尔控制变异数方法在增加运算效率方面是非常有用的，并且也保留了运算的正确性。

A: NVIDIA's Physx technology is the use of massive parallel processing GPU threads physical framework for graphical real-time changes in vector calculus, this is a good idea, you can significantly reduce the burden on CPU, allowing more gaming performance by significantly cards to decide.

答：NVIDIA的Physx技术是利用GPU庞大的并行处理线程来为图形物理框架的变化矢量做即时演算，这的确是个好想法，可以大幅度减轻CPU的负担，可以让游戏性能更多地由显卡来决定。

In addition, since the cloth simulation algorithm requires complex mathematics and processes large matrices of data, we were able to leverage the vector support of the Cell processor and achieve even greater performance by vectorizing the code that performs arithmetic functions on the mesh data structures.

因为，布料模拟数据量巨大，我们设法提高了对cell的向量支持能力，并且通过向量化负责演算网格数据结构的函数代码来获得性能得巨大提升。

Weka如下:Weka is a collection of machine learning algorithms for data mining tasks.

WEKA是一个开放的平台，内有许多常用的资料探勘的演算法及方法，由於是开放的平台，使用者也可以自行撰写自己的演算法。

This thesis studies a widely used first order optimzation algorithm--Sequential Linear Programming method, SLP is modify so that it can solve structural and shape optimization problems more efficiently, including continuous and dsicrete variables problems. Finally the zeroth order optimization method is presented for two dimensional shape optimization problems.

本研究将从一般广为设计者所采用的一次最佳化方法--线性序列规划法的修正研究，以使其更稳定且更有效率地解决在一般结构与形态最佳化问题，包含有连续式及非连续式设计变数的最佳化问题演算过程，最后应用零次最佳化方法概念在二维形态最佳化设计上。

The results indicate that heuristic algorithm can reasonably be applied to identify zonation of groundwater parameters.

本研究成果证明敔发式演算法可合理应用於地下水参数分区。

One algorithm used a modified COMO's matrix-clustering algorithm [9], while the other is average-linkage agglomerative clustering based on the similarity [10] among classes as a measure of distance.

其中一种演算法是运用修正过后COMO模式的矩阵群集演算法[9]，而另一种是平均联结凝聚的群集方法，其距离是依类别之间的相似度做为量测基础[10]。

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。