Experience-consistent modeling: Regression and classification problems

In this study, we are concerned with system modeling which involves limited data and reconciles the developed model with some previously acquired domain knowledge being captured in the format of already constructed models. Each of these previously available models was formed on a basis of extensive data sets which are not available for the current identification pursuits. To emphasize the nature of modeling being guided by the reconciliation mechanisms, we refer to this mode of identification as experience-consistent modeling. The paper presents the conceptual and algorithmic framework by focusing on regression models. By forming a certain extended form of the performance index, it is shown that the domain knowledge captured by regression models can play a similar role as a regularization component used quite commonly in system identification. Experimental results involve both synthetic low-dimensional data and selected data coming from Machine Learning repository. The data used in the experiments tackle regression models as well as classification problems (two-class classifiers).     

Some First-Order Algorithms for Total Variation Based Image Restoration

This paper deals with first-order numerical schemes for image restoration. These schemes rely on a duality-based algorithm proposed in 1979 by Bermùdez and Moreno. This is an old and forgotten algorithm that is revealed wider than recent schemes (such as the Chambolle projection algorithm) and able to improve contemporary schemes. Total variation regularization and smoothed total variation regularization are investigated. Algorithms are presented for such regularizations in image restoration. We prove the convergence of all the proposed schemes. We illustrate our study with numerous numerical examples. We make some comparisons with a class of efficient algorithms (proved to be optimal among first-order numerical schemes) recently introduced by Y. Nesterov.     

基于流记录的主干网活跃IP地址空间检测

掌握IP地址的实际使用情况对于网络管理和网络安全等研究领域有着重要的意义.提出一种以抽样流记录为分析数据源的活跃地址检测算法,其核心思路是将存在双向通信流量作为地址活跃判定条件.算法基于被动测量技术,以流记录为分析数据源使其可以在主干网边界运行.讨论了抽样、伪造地址等问题对算法的影响以及相应的应对策略,用DPI分析检验了算法的准确性和有效性.最后基于NBOS平台,将其部署在CERNET全部38个主节点,完成了全网活跃IP地址空间的检测.     

代码/需求行为差异检测

为解决软件开发后期（维护/演化）程序代码与需求模型不一致的问题,本文面向逆向需求工程,重点研究检测变更代码与原始需求模型之间行为差异的算法：首先沿用模型/代码转换技术,分析模型/代码比较原理,设计比早期连续型单向串行检测算法快2N倍（N为路径数）的离散型双向并行检测算法。然后采用该算法开发图形化需求/代码比较工具RCCT,并将其集成进综合需求建模系统（RMTS）,使动画建模、特性检测、模型转换、需求/代码差异检测等功能融为一体。最后,通过电子转账案例演示该工具的使用方法,并编写测试程序证明离散型双向并行算法不但比原始算法高效,而且更加可靠。     

新型光滑正则半监督SVM方法及其在信用评级中的应用

提出了一种基于光滑正则的半监督支持向量机方法,并将其用于建立中小信用评级模型.它从少量标签样本和大量无标签样本中构造反映数据流形结构的光滑正则项,并结合到支持向量机的最大间隔分类器的学习过程.然后,提出一种渐进式方法来迭代获得“半标签”样本,稳健地提升支持向量机的泛化性能.在真实数据集上的实验结果表明,新方法获得的测试精度显著优于多种现有方法,非常适用于中小企业的信用评级任务.     

弹性多核学习

多核学习 (MKL) 的提出是为了解决多个核矩阵的融合问题, 多核学习求解关于多个核矩阵的最优的线性组合并同时解出对应于这个组合矩阵的支持向量机(SVM)问题. 现有的多核学习的框架倾向于寻找稀疏的组合系数, 但是当有信息的核的比例较高的时候, 对稀疏性的倾向会使得只有少量的核被选中而损失相当的分类信息. 在本文中, 我们提出了弹性多核学习的框架来实现自适应的多核学习. 弹性多核学习的框架利用了一个混合正则化函数来均衡稀疏性和非稀疏性, 多核学习和支持向量机问题都可以视作弹性多核学习的特殊情形. 基于针对多核学习的梯度下降法, 我们导出了针对弹性多核学习的梯度下降法. 仿真数据的结果显示了弹性多核学习方法相对多核学习和支持向量机的优势; 我们还进一步将弹性多核学习应用于基因集合分析问题并取得了有意义的结果; 最后, 我们比较研究了弹性多核学习与另一种利用了非稀疏思想的多核学习.     

Frequency domain regularization of d-dimensional structure tensor-based directional fields

The present work is intended to address two of the major difficulties that can be found when tackling the estimation of the local orientation of the data in a scene, a task which is usually accomplished by means of the computation of the structure tensor-based directional field. On one hand, the orientation information only exists in the non-homogeneous regions of the dataset, while it is zero in the areas where the gradient (i.e. the first-order intensity variation) remains constant. Due to this lack of information, there are many cases in which the overall shape of the represented objects cannot be precisely inferred from the directional field. On the other hand, the orientation estimation is highly dependent on the particular choice of the averaging window used for its computation (since a collection of neighboring gradient vectors is needed to obtain a dominant orientation), typically resulting in vector fields which vary from very irregular (thus yielding a noisy estimation) to very uniform (but at the expense of a loss of angular resolution). The proposed solution to both drawbacks is the regularization of the directional field; this process extends smoothly the previously computed vectors to the whole dataset while preserving the angular information of relevant structures. With this purpose, the paper introduces a suitable mathematical framework and deals with the d-dimensional variational formulation which is derived from it. The proposed formulation is finally translated into the frequency domain in order to obtain an increase of insight on the regularization problem, which can be understood as a low-pass filtering of the directional field. The frequency domain point of view also allows for an efficient implementation of the resulting iterative algorithm. Simulation experiments involving datasets of different dimensionality prove the validity of the theoretical approach.     

基于特征的自适应正则化配准算法*

Andriy Myronenko提出了一种自适应正则化的方法并将其应用于非刚性图像的配准,该方法在配准速度和配准精确度方面都取得了比较好的效果。但该方法对变形场初始值比较敏感,选择不当则会陷入局部极小值而不能得到理想的配准结果。为了使原始算法得到更广泛的应用,本文引入了基于特征点的粗配准方法,得到了与真实变形场更加接近的初始变形场,从而摆脱了局部极小值的困扰,得到了正确的配准结果。实验证明,改进后的算法在应用范围和配准精度上都有了提高。     

图像复原的Contourlet收缩与广义全变分正则化方法

建立了广义全变分(total variation,TV)模型,分析正则项在复原算法中的作用.分别从图像的平坦区域和边缘区域入手,在平坦区域图像各向同性扩散,在边缘区域则要满足各向异性扩散,从理论上对两种情形下的扩散做深入分析,推导出广义TV模型满足的一些条件,为了防止高噪声情形下复原模型失效以及克服方块效应,在正则项中引入了 Contourlet收缩,它是一种多分辨的、局域的、多方向的更稀疏的图像表示方法,正则项中引入的Contourlet收缩具有去噪和提取图像重要信息的作用,Contourlet收缩与广义TV正则化相结合,兼顾了图像的光滑性和边缘保持,特别是在图像严重模糊、噪声越多的情形下,更加体现了这种算法比改进的TV模型有效.