查询词典 structure function

The relationship between the topology and convegence of networks is approached and the reason causing overtraining and overfitting is analysed. Then the redundant weights and nodes in the network are defined by using improved activity function and a new algorithm called the BPNSO for determining the network structure for any application is proposed, which use dual optimization objective function combining the mean squre error of the network output with network complexity penalty function.

文章对网络拓扑结构与网络收敛性和稳定性的关系进行了探讨，分析了网络出现Overtraining或Overfitting的原因和解决的目标，提出了神经网络结构优化的策略，改进了活化函数，定义了冗余权值和冗余节点的概念，为辨识和删除冗余权值和冗余节点，提出采用优化目标函数中增加网络复杂度罚函数的双重优化目标函数，并对罚因子提出了基于学习误差和权值的自适应动态调整公式，论文定义了这种BP网络结构优化方法为BPNSO，并给出了完整的计算流程。

In this article, the domain structure of LBD genes, classes of LBD genes in model plants of monocotyledon or dicot, expression model, mutation phenotypes from loss of function or gain of function, and its relationship to other gene families were reviewed. Furthermore, the redundancy character of LBD genes, basic candidate function shared by LOB domain, and potential molecular mechanism of LBD genes interacting with other genes also were discussed.

本文对LBD基因的结构域特征、在单／双子叶模式植物中的分类、表达特点、己克隆LBD基因的功能及与其它基因或基因家族间的相互作用关系进行了综述，并对LBD基因在高等植物中的功能冗余特性、LOB结构域可能所具有的基本功能和LBD基因与其它基因相互作用的分子机制进行了探讨。

It solves the problem that the unitary contour presentation can not correctly extract face contour in a face image which suffers from scale, rotation etc. The definition of the internal and external energy function is provided. At the same time, the global matching algorithm and local matching algorithm is given. The experiment shows that this presentation and the accompanying matching algorithm can be used to extract the face contour very well. So the image segmentation can be implemented by using it.②By analyzing the recognition principle of PCA method, we can conclude that the face images coming from different surrounding consist of different face image space. This is the essential reason that makes the generality of PCA method worse. Also, we give a measurement means to measure the distance from different face image space, so we can analyze face image space more conveniently.③We also construct various scale models and rotation pose models to detect the scale and rotating angle of face image to be recognized. The experiment results show that the detecting precision is very high. So it is good for face image feature extraction and face image representation.④Similarly, we construct local feature models of face image and utilize them to detect the local feature of face image. At the same time, we put forward a novel face image local feature detection algorithm, locating step by step. The experiment results show that this method can accurately detect the location of local face feature in a image.⑤A novel face image presentation model, dual attribute graph , is put forward. Firstly, it utilizes attribute graph to present the face image, then exact the local principal component coefficient and Gabor transform coefficient of thc pixels which corresponds to the nodes of the graph as the attribute of the nodes. This representation fully makes use of the statistical characteristic of the local face feature and utilizes Gabor transform to present the topographical structure of face image. So DAG has more general property.⑥Based on the DAG presentation, we give a DAG matching function and matching algorithm. During the design of the function and algorithm, the noise factor, e. g., lighting, scale and rotation pose are considered and tried to be eliminated. So the algorithm can give more general property.⑦A general face image recognition system is implemented. The experiment show the system can get better recognition performance under the noise surrounding of lighting, scale and rotation pose.

本文在上述研究的基础上，取得了如下主要研究成果：①构造了一个通用的人脸轮廓模型表示，解决了由于人脸图象尺度、旋转等因素而使得仅用单一轮廓表示无法正确提取人脸轮廓的问题，并给出了模型内、外能函数的定义，同时给出了模型的全局与局部匹配算法，实验表明，使用这种表示形式以及匹配算法，能够较好地提取人脸图象的轮廓，可实际用于人脸图象的分割；②深入分析了PCA方法的识别机制，得出不同成象条件下的人脸图象构成不同的人脸图象空间的结论，同时指出这也是造成PCA方法通用性较差的本质原因，并给出了不同人脸空间距离的一种度量方法，使用该度量方法能够直观地对人脸图象空间进行分析；③构造了各种尺度模板、旋转姿势模板以用于探测待识人脸图象的尺度、旋转角度，实验结果表明，探测精确度很高，从而有利于人脸图象特征提取，以及图象的有效表示；④构造了人脸图象的各局部特征模板，用于人脸图象局部特征的探测；同时提出了一种新的人脸图象局部特征探测法---逐步求精定位法，实验结果表明，使用这种方法能够精确地得到人脸图象各局部特征的位置；⑤提出了一种新的人脸图象表示法---双属性图表示法；利用属性图来表示人脸图象，并提取图节点对应图象位置的局部主成分特征系数以及Gabor变换系数作为图节点的属性，这种表示方法充分利用了人脸图象的局部特征的统计特性，并且使用Gabor变换来反映人脸图象的拓扑结构，从而使得双属性图表示法具有较强的通用性；⑥在双属性图表示的基础上，给出双属性图匹配函数及匹配算法，在函数及算法设计过程中，考虑并解决了光照、尺度、旋转姿势变化等因素对人脸图象识别的影响，使得匹配算法具有较强的通用性；⑦设计并实现了一个通用的人脸图象识别系统，实验结果表明，该系统在图象光照、尺度、旋转姿势情况下，得到了较好的识别效果。

The fractional structure of the objective function maximizing RAROC makes it difficult to be solved. We noticed that in real financial practice, the expected return function and risk function of the portfolio are often homogeneous with the amount of investment.

根据RAROC的分式结构，以及回报函数和风险函数通常是关于投资额的齐次函数，将分式优化问题转化为对其分母的最优化。

In chapter 3 constructs firstlya new differential merit function by a perturbation structure of VIP and study theproperties of this merit function.Based on this merit function,a continuation-typeNewton method is proposed,which use the technique of inexact linear search forassuring its global convergence,and improves essentially the method of Taji,Fukushima and Ibaraki which can only solve the strong monotone VIP,and has locallyquadratic convergent rate under some condition.

第三章利用变分不等式问题的一种扰动结构构造了新的可微效用函数，并研究了效用函数的性质；在此基础上，给出了一类求解一般单调变分不等式问题的连续型Newton方法CN，方法采用了不精确的线性搜索技术以确保整体收敛性，从本质上改进了Taji、Fukushima和Ibaraki方法只能求解强单调变分不等式问题的局限性，同时又保持了局部二次收敛率。

The absolute solution of Jiaojia ERP is put forward, the system models are established, and the solution are explained in detail. The business program of Jiaojia gold mine is reengineered based on the theory of "business centers", and the program-oriented compressed organization is framed. The index group and the solution of mine total quality management are setup. Then the optimize program of SCM and its components are designed on the guidance of "Horizontal Integration". The data of Jiaojia gold mine are reengineered based on the data warehouse technology. The information integration can be available after the establish of the business oriented databases, the subject integration oriented data marts, the application oriented data warehouses, the code databases and metadata, as weil as the OLAP system. The mathematics models and data mining modes are set up to function for the decision support of ERP system. The model of programming Master Production Schedule driven by geological resources and marked is figured out, so as to the models for material requirements planning and other decision supports are set up. The models provide service in forms of model storehouse, method storehouse, and algorithm tools center. The agent center is introduced to process the models' organizing, service, as well as self-study. The function system of Jiaojia gold mine ERP is figured out. By combining the development characteristics of Clients/Server and Browse/Server, the general framework and every sub systems are described in detail, including the function, structure, arrangement, modality, and so on.

主要内容如下：（1）深入论述了黄金矿山实施ERP的可行性，并结合应用需求将ERP理论在深度和广度上进行了拓延，建立了黄金矿山ERP系统的理论基础：以资源与市场共同驱动的供应链管理；黄金矿山的双重物流体系；以地质资源有效利用为目标的生产控制体系；黄金矿山财务管理的成本控制核心理论等；（2）提出焦家金矿ERP系统的整体解决方案，构建ERP系统模型，并进行了细节方案的设计：基于"作业过程核心"理论重新规划焦家金矿的业务流程，在此基础上设计面向流程的"扁平化"组织机构；建立了矿石全面质量管理的指标体系和实施方案；以"横向一体化"为指导思想，提出供应链管理各组成部分的全面优化方案；（3）以数据仓库技术为核心完成焦家金矿的数据规划，分别建立面向业务的数据库、面向主题集成的数据集市和面向主题应用的数据仓库，设计代码数据库和元数据库，定义OLAP方式，实现焦家金矿的全面信息集成；（4）针对ERP系统中的决策支持功能，建立了相应的数学模型和数据挖掘模型，提出了地质资源和市场共同驱动的主生产计划模型制定过程，进一步建立了物料需求计划的线性规划模型，以及其它决策支持模型；提出以模型库、方法库和算法工具为核心的模型服务方式，并设计了代理机制完成模型的组织、服务和自学习过程；（5）构建焦家金矿ERP系统的功能体系，针对C/S和B/S模式系统的开发特点，分别对功能体系架构以及各子系统的功能、结构、层次和形式进行详细设计。

Reduced SQP based on orthonormal bases, orthogonal bases and coordinate bases are studied and compared. With the correction item introduced, the null-space move is adjusted and the reduced SQP based on coordinate bases decomposition is improved.●The key problem of steady-state data processing for RTOPT is studied. A new algorithm of simultaneous strategies for data reconciliation and gross error detection is proposed. Here the objective function that is constructed using maximum likelihood principle and joint distribution function is minimized. Efficient algorithm is developed which exploited the nonlinear lease-square structure of this objective function.

仿真结果表明了算法的有效性；●深入研究了在线优化中的一个关键问题——稳态数据处理问题，用基于组合概率分布的目标函数联立求解粗劣误差检测和数据调和问题；以精馏塔在线优化的数据处理和模型修正为例，提出一种根据模型方程对参数的依赖性进行过程模型分解的分层策略，可以将数据调和和参数估计统一在数据处理的大框架中，为进一步的过程性能分析、操作状况评价以及在线优化提供可靠的基准，具有非常重要的工程应用价值；●将优化命题的形式扩展到更通用的形式。

Secondly, based on the different structure characteristics and additional conditions, we study several kinds of inverse problems of pseudoparabolic equations. One is a kind of pseudoparabolic inverse problem of identifying a constant coefficient solved by combining the formal solution of the problem and the additional condition properly. The second is the pseudoparabolic inverse problems of identifying an unknown boundary function and an unknown source term solved by using the Riemann function method to get the formal solution of the problem and then using the additional condition to transform the problem into a Volterra integral equation of the second kind. The third is a kind of backward heat flow problem of nonlinear pseudoparabolic equation solved by combining the Riemann function method and the fixed point theory properly.

其次，根据不同模型的结构特点和附加条件，研究了几类伪抛物型方程的反问题：一是利用问题的形式解并结合附加条件，解决了一类伪抛物型方程常数系数的反问题；二是利用Riemann函数方法获得问题的形式解，利用附加条件将问题转化成求解第二类Volterra积分方程问题，解决了一类伪抛物型方程未知边界值的反问题和未知源项的反问题；三是将Riemann函数方法和不动点定理相结合，解决了一类非线性伪抛物型方程的后向热流问题。

Results of analysis show that Generalized Paris Law can consider the complex crack propagation in the material of the pavement structure. The fatigue life of the pavement increases with increasing thickness of surface layer in a power function. Use of a thicker surface layer may extend the service life of the asphalt pavement. The fatigue life increases with decreasing modulus of surface layer in a power function with a negative exponent. However, since reduced modulus would scarify the structural strength and might cause rutting-related distress, caution should be taken when using low modulus materials for improving fatigue performance of asphalt pavements. The fatigue life increases with increasing modulus of base material in a power function. However, the fatigue life can not infinitely increases with increasing modulus of the base.

计算与分析表明：广义Paris公式可以考虑沥青路面结构内材料复杂的裂缝扩展：沥青面层的疲劳寿命随着面层厚度的增加以幂函数的形式增加，适当增加沥青面层厚度可以提高路面的使用寿命；沥青面层的疲劳寿命随着面层材料模量的降低以负指数的幂函数形式增加；由于材料模量的降低将牺牲路面的整体强度并可能引起路面车辙类损坏，以此提高疲劳寿命的方法应慎重对待；沥青面层的疲劳寿命随着基层材料模量的增加呈幂函数的形式增加，但由疲劳方程可以看出，寿命并不是随基层材料模量的增加而无限的增加，疲劳曲线在经过一个上升段后，逐渐趋于一个常数值，这说明，这时基层模量对寿命已不作关键性贡献；随着底基层厚度的增加，面层的疲劳寿命近似地呈线性函数增加，但效果并不很明显。

The client application is designed under multiple document interface structure to make it both aesthetic and practical. The task of security management may be done at the application level with the aid of safe logon function module. The data query function is very applicable and could be applied in many cases. Graph section displays heating information with diverse kinds of figures. Two completely different printing ways are depicted in the printing function module.

客户端应用程序的设计采用了多文档结构，力求美观、实用；系统安全登录功能的设计使安全管理工作可以在应用程序的层次上完成；数据的查询操作适用性强，并具有一定的通用性；图形部分实现了供热信息的多种图形显示；打印处理模块描述了两种不同的数据打印处理方式。

Many will continue to choose to live in duality and in conflict.

许多人将继续选择活在二元对立性和冲突中。

I find that students of the University of Physical Education all wear sportswear at first sight.

我发现：体育大学的学生乍一看，都是穿运动衣，大家都一样