查询词典 continuation method

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。

A mixed method of conjugate gradient method and steepest descent method ;2. The Modified Steepest Descent Method──Best Point in Steepest Descent method;3. This paper presents the mathematical model for the optimization of heterogeneous components, and the method using sensitivity analysis and steepest descent method to optimize material properties, the component is then identified.

阐述了非均质材料零件设计优化的数学模型，并采用灵敏度分析以及最速下降法对其各个材料区域的材料性能进行设计优化，得到最佳材料性能参数后，再从非均质材料数据库中找到相应的工程材料，合成满足设计要求的非均质材料零件该方法为设计者提供了切实可行的非均质材料零件的材料设计方

By using accelerating Aitken method in iterative method of nonlinear equations, a new modified method is given. The method is applied to modified Newton method, thereby a new method is developed, which is faster and stable. Also, its algorithm is given. Numerical results show that the new method is very efficient.

通过对非线性方程组的迭代法引入Aitken加速技术，设计了一种非线性方程组的迭代解法的加速方法；将该方法与M步Newton法相结合，得到一种收敛速度快而且计算稳定的方法，并给出了具体算法；数值结果表明了新算法是有效的。

On the design of algorithms, a novel exact hierarchical delay analysis method for general circuits is proposed; based on the sensitization theorem for sequential circuits, an exact minimizing clocking method is proposed; based on Boolean process, a waveform simulation method considering interconnecting delay for logic circuit and a parallel waveform simulation method are proposed; a new method that transforms bit-level waveform polynomial to word-level polynomial model is proposed; a multiple valued synthesis algorithm based on multiple valued Boolean process and a wire-centered delay synthesis policy are proposed, in which timing planning, floorplanning, wire planning and optimal clock skew in early design are considered; a two-layers channel routing method for minimizing crosstalk under grid mode is proposed; based on the transition numbers theorems for waveform polynomial, a new method for generation of test with noise effects is proposed.

算法设计方面，提出了一种精确的通用电路层次化延时分析方法；基于时序电路的敏化定理提出时序电路最小时钟周期精确确定方法；提出基于Boolean过程论的考虑互连延迟的逻辑电路波形模拟方法，在分析了波形模拟适合并行化基础上，进一步提出一种并行波形模拟算法；提出一种将位级电路波形多项式描述转化成字级多项式描述的新方法；提出一种基于多值Boolean过程的多值电路综合算法以及一种将前期设计定时规划、前期设计的布局规划和线网结构化方法及低偏移的时钟分配等技术相结合的面向互连延时的综合策略；提出一种串绕最小化的网格模式下的双层通道布线方法；从波形多项式描述跳变数的定理出发提出了一种考虑噪声效应的测试生成新方法。

There are many methods of foundation consolidation for civil engineering, such as soil exchange method, prepress method, dynamic consolidation method, vibrancy rushing method, soil and podsol dense pile method, sand pile, cement-coal-powder and gravel pile method, deep mixing method, high-pressure eject masonry, etc.

目前国内外地基处理的方法很多，主要的地基处理方法包括：换填法、预压法、强夯法、振冲法、土和灰土挤密桩法、砂桩法、水泥粉煤灰碎石桩法、深层搅拌法以及高压喷射注浆法等。

To date, many investigations have focused on the ceramic fibers, and some methods such as melt/drawing method, slurry method, solution dipping method, CVDmethod, CVR method, sol-gel method, preceramic polymer method, hydrothermal method have been developed for the preparation of the ceramic fibers.

各种制备方法都各有优缺点，各种制备技术因所制备的纤维性能和用途不同而使其依然发挥各自不同的作用，但纵观连续陶瓷纤维制备技术的发展轨迹，可以发现溶胶-凝胶法日益成为研究与开发的主流。

Article 12 Commercial banks should explain in their notes to the accounting report the important policy of accounting and accounting estimates, including: Accounting standards, accounting year, reporting currency, accounting basis and valuation principles; Type and scope of loans; Accounting rules for investment; Scope and method of provisions against asset losses; Principle and method of income recognition; Valuation method for financial derivatives; Conversion method for foreign currency business and accounting report; Preparation method for consolidated accounting report; Valuation and depreciation method for fixed assets; Valuation method and amortization policy for intangible assets; Amortization policy for long-term deferred expenses; Accounting practice for income tax.

第十二条商业银行应在会计报表附注中说明本行的重要会计政策和会计估计，包括：会计报表编制所依据的会计准则、会计年度、记账本位币、记账基础和计价原则；贷款的种类和范围；投资核算方法；计提各项资产减值准备的范围和方法；收入确认原则和方法；衍生金融工具的计价方法；外币业务和报表折算方法；合并会计报表的编制方法；固定资产计价和折旧方法；无形资产计价及摊销政策；长期待摊费用的摊销政策；所得税的会计处理方法等。

Applying this method, Interpolation Method and Comprehensive Trophic Level Index Method to evaluate eutrophication of Chaohu Lake, the results showed that there was significant difference between this method and Interpolation Scoring Method, and the most significant difference between this method and Comprehensive Trophic Level Index Method.

该方法对巢湖富营养化的综合评价结果与插值评分法有显著差异，与综合营养状态指数法有极显著差异。

In order to make the new recursive method convergence fast, the RLS-like forgetting factor is introduced to construct the new data Hankel matrices. In the framework of subspace methods, the gradient type subspace tracking method is used to implement the update of the state subspace, which results the unbiased estimate. Different from traditional offline method, the system matrices are recursively computed using RLS method. The convergence analysis of the proposed method is given and finally, the efficiency of this method is illustrated with a simulation example.A new closed-loop identification method is proposed to implement consistency estimation of closed-loop subspace identification.

引入RLS-like遗忘因子机制构造新的数据矩阵，更好地跟踪系统时变信息，提高递推算法的收敛速度；在子空间辨识框架内，采用梯度型子空间跟踪算法实现对状态子空间的更新，避免了子空间近似带来的估计有偏性；与传统的离线算法不同，在得到系统状态子空间后，基于递推最小二乘方法提出新的算法，实现对系统矩阵的递推估计；并给出递推子空间辨识算法的收敛性分析；最后通过仿真实例验证算法的有效性。

Using the one-atom theory, the electronic structures of Pt-electrocatalyst with fcc structure was determined as follows:(5dn)6.48 (5dc)2.02 (6sc)1.48(6sf)0.02. Compared to the result obtained by the first-principle method such as FP-LMTO, CASTEP, their results are in good agreement with each other. Comparing the calculations of physical properties such as lattice constants, cohesive energy and bulk modulus by OA method and first-principle method, the result obtained by OA method is in excellent agreement with experimental value, but the result obtained by first-principle method is not accordable. The relationship between the electronic structures and catalytic performance was explained qualitatively by OA method and first principle method according to its electronic structures. Because the d-orbital vacancies increases and static density is high around Fermi energy band, Pt has good catalytic performance.

摘 要：依据纯金属单原子理论确定面心立方结构电催化剂Pt的电子结构为(5dn)6.48(5dc)2.02-(6sc)1.48(6sf)0.02，与采用第一原理的FP-LMTO和CASTEP等方法所计算的电子结构相比较，其结果非常相近；由OA理论和第一原理方法计算的晶格常数、结合能、体弹性模量等物理参数进行比较，OA理论计算的结果与实验值较符合，而第一原理方法计算的结果与实验值相差较大；在此基础上用OA理论和第一原理方法研究了Pt的电子结构与催化性能的关系，由于d带空穴增多和费米能级附近态密度较高，导致金属Pt的催化性能很好。

You can do some assertiveness training.

你可以进行一些自信训练。

We were well on the path to making a rear-wheel-drive global platform," says Mays."

我们正致力于建立一个后轮驱动的平台，"Mays这样说道。"