Interactive search for the global maximum of a function of many variables

This paper discusses a new algorithm to locate the global maximum of a function defined in a multidimensional rectangular domain. The number of dimensions is as large as, or even more than, 5 or 10. There are two important elements in this algorithm. One is the transformation of the object function in such a way that its global maximum corresponds to infinity while other secondary maxima are reduced to zero. Actually there is some departure from the ideal transformation because of possible overflow on the computer. This portion of the algorithm precedes the interactive (or conversational) use of a graphic display system. This interactive part makes the other element of the algorithm. A multidimensional point is represented as a curve on the display screen. By projecting numerous points in the multidimensional space to similarly numerous curves on the screen of the graphic display device, the human eye can make overall recognition much more efficiently than computers. This fact is exploited to reduce the problem to that of a set of unimodal peaks. Once the supporting domain for each of these peaks is separated by visual aid, one may leave the computer to handle the rest of the problem for itself. A number of numerical experiments are done and discussed to provide evidence regarding the feasibility of the proposed algorithm.     

Directional interpolation via all-pass transfer function matricesand its application in Hankel-norm approximations

A new approach to the problem of multivariable interpolation via all-pass transfer function matrices that are not necessarily stable is presented. The approach applies to both state-space and classical function theoretic arguments and obtains a very simple expression for the all-pass matrix that satisfies the interpolation requirement. Unlike the solution that is obtained by the generalized Nevanlinna-Pick algorithm, this expression is derived in closed form explicitly in terms of the interpolation parameters. It allows a detailed investigation of the structure of the all-pass solution, and it is readily used in Hankel-norm approximations of linear multivariable systems     

Augmenting learning function to Bayesian network inferences with maximum likelihood parameters

Computing the posterior probability distribution for a set of query variables by search result is an important task of inferences with a Bayesian network. Starting from real applications, it is also necessary to make inferences when the evidence is not contained in training data. In this paper, we are to augment the learning function to Bayesian network inferences, and extend the classical “search”-based inferences to “search + learning”-based inferences. Based on the support vector machine, we use a class of hyperplanes to construct the hypothesis space. Then we use the method of solving an optimal hyperplane to find a maximum likelihood hypothesis for the value not contained in training data. Further, we give a convergent Gibbs sampling algorithm for approximate probabilistic inference with the presence of maximum likelihood parameters. Preliminary experiments show the feasibility of our proposed methods.     

Computational aspects of maximum likelihood estimation and reduction in sensitivity function calculations

This paper discusses numerical aspects of computing maximum likelihood estimates for linear dynamical systems in state-vector form. Different gradient-based nonlinear programming methods are discussed in a unified framework and their applicability to maximum likelihood estimation is examined. The problems due to singular Hessian or singular information matrix that are common in practice are discussed in detail and methods for their solution are proposed. New results on the calculation of state sensitivity functions via reduced order models are given. Several methods for speeding convergence and reducing computation time are also discussed.     

方向角模拟器设计

方向角信息是飞行参数信号中比较重要的记录对象，因此，通过模拟方向角来测试飞行参数的记录仪对于方向角能否被正常记录及记录精度尤为重要。本文介绍了方向角模拟器各模块的组成、结构及原理，并对影响模拟方向角信号误差的因素进行了理论分析，为减小误差提供了理论依据。通过减少系统整流电路的误差和修正A／D转换的非线性误差，从而减少了系统的整体误差，获得了比较高的精度。     

Directional filling-in

The filling-in theory of brightness perception has gained much attention recently owing to the success of vision models. However, the theory and its instantiations have suffered from incorrectly dealing with transitive brightness relations. This paper describes an advance in the filling-in theory that overcomes the problem. The advance is incorporated into the BCS/FCS neural network model, which allows it, for the first time, to account for all of Arend's test stimuli for assessing brightness perception models. The theory also suggests a new teleology for parallel ON- and OFF-channels.     

基于p阶最大相关熵准则损失函数的鲁棒极限学习机

当前对于短期负荷预测的研究主要针对影响因素的分析以及模型的改进,很少有对模型的鲁棒性进行研究.以极限学习机(extreme learning machine,ELM)作为研究对象,针对ELM模型的鲁棒性问题进行了深入的研究,并将其应用到短期负荷预测问题中.ELM模型的鲁棒性受损失函数的影响,当前ELM模型在处理含异常点样本时,鲁棒性差、预测精度较低.针对该问题,提出了一种基于p阶最大相关熵准则的损失函数,并将该损失函数应用到ELM模型中,以提高其在短期负荷预测问题中的鲁棒性.提出了一种估计实际样本中异常点百分比的计算方法,在建立短期负荷预测模型之前,估计出实际负荷样本中的异常点百分比.仿真结果表明,在异常点超过12％的样本中,提出的算法模型具有更好的鲁棒性以及预测精度.     

方向性小波理论应用特性分析

方向性小波理论为图像处理提供了一种新的稀疏表示方法,能够更有效地捕捉图像中的几何结构。本文从基的特征入手,比较了方向性小波与传统小波在逼近图像几何边缘时的不同之处;总结了近年来该领域内提出的几种主要理论,以Ridgelet变换为例说明了方向性小波理论的基本原理。实验演示了Contourlet变换和小波变换的非线性逼近性能
能和去噪效果。最后指出了该领域进一步研究的方向     

Development of the Omni Directional Intelligent Navigator

The Autonomous Systems Laboratory is in the midst of developing an advanced underwater robotic technology test platform. The platform consists of the Omni-Directional Intelligent Navigator (ODIN) and the Integrated Graphic Workstation (IGW). ODIN is a six degree-of-freedom (dof) underwater vehicle with dual operational modes (autonomous and tethered) and a single dof mechanical manipulator. IGW is a real-time, 3-dimensional graphic monitoring, testing, and evaluation workstation. This paper presents ODIN's mechanical and electrical specifications; its vehicle dynamics and depth control system; its recent simulation and experimental results; and IGW's specifications     

Directional interpolation in the state space

This paper presents a new state-space characterization of the directional interpolation based on the observation that the generalized Pick matrix is regarded as a solution of a Lyapunov-type equation. A simple representation of the set of all solutions is derived in terms of linear fractional transformations. The computation of the associated J-unitary matrix is simpler than the well-known procedure based on the Glover's scheme or the J-spectral factorization of Ball and Ran.     

Directional morphological filtering

We show that a translation invariant implementation of min/max filters along a line segment of slope in the form of an irreducible fraction dy/dx can be achieved at the cost of 2+k min/max comparisons per image pixel, where k=max(|dx|,|dy|). Therefore, for a given slope, the computation time is constant and independent of the length of the line segment. We then present the notion of periodic moving histogram algorithm. This allows for a similar performance to be achieved in the more general case of rank filters and rank-based morphological filters. Applications to the filtering of thin nets and computation of both granulometries and orientation fields are detailed. Finally, two extensions are developed. The first deals with the decomposition of discrete disks and arbitrarily oriented discrete rectangles, while the second concerns min/max filters along gray tone periodic line segments     

极大熵和声搜索算法求解多目标优化*

给出了求解多目标优化问题的一个新算法。首先利用极大熵函数,将多目标优化问题转换为一个单目标优化问题;然后利用和声搜索算法对其进行求解,进而得到多目标优化问题的有效解。该算法对目标函数的解析性质没有要求且容易实现,数值结果表明了该方法是有效的。     

Estimating the maximum information leakage

Preventing improper information leaks is a greatest challenge of the modern society. In this paper, we present a technique for measuring the ability of several families of adversaries to set up a covert channel. Our approach relies on a noninterference based formulation of security which can be naturally expressed by semantic models of the program execution. In our analysis the most powerful adversary is measured via a notion of approximate process equivalence. Even if finding the most powerful adversary is in general impractical, we show that this requires only a finite number of checks for a particular family of adversaries which are related to a probabilistic information flow property.

基于定向天线的飞行自组网定向路由协议综述

近年来,以无人机为节点的飞行自组网因其在各个领域的不同应用而受到广泛关注.为满足复杂任务的服务质量需求,飞行自组网的路由需要提供足够高的网络性能.相比于基于全向天线的全向路由,基于定向天线的定向路由能提升信道利用率,扩大通信范围,可以使飞行自组网获得更好的网络性能和服务质量.文中综述了基于定向天线的飞行自组网定向路由,分析了在飞行自组网中应用定向天线的优势和所带来的问题,而后对现有单路径定向路由和多路径定向路由从定向天线控制机制、路由算法、使用场景和优缺点等多个方面进行详细介绍,并从天线类型、控制机制、网络性能和关键参数等多方面对这些路由协议进行定性比较,最后讨论了基于定向天线的飞行自组网定向路由协议在实际应用中和未来发展中所面临的挑战.     

方向关系的定性表示与推理

方向关系是GIS中的一个重要的理论问题。本文重点讨论方向关系的定性表示与推理。文章首先讨论了基于井字空间的方向关系定性表示,然后介绍了基于方向关系组合运算表的方向关系定性推理,最后对我们现在和将来的一些研究进行了叙述,并对我们的工作做了总结。     

试论动画素描教学的专业指向性

动画专业是培养动画艺术作品设计与制作专门人才的教学学科,具有较强的专业性。作为这一专业的基础课程,动画素描的教学必然在一定程度上相应地体现出与传统绘画素描和现代设计素描不同的、与动画专业特点相对应的明显的专业指向性。因而,如何构建与动画专业教学相适应的素描教学体系,从而使动画素描教学适应动画专业教学、真正为动画专业教学与动画创作奠定基础,是当前动画专业教学亟待解决的问题,是一个值得我们去探讨的课题。     

指纹方向图的改进研究

提出一种在现行方向图的基础上改进的圆形方向图,这种圆形方向图过渡平滑、自然、真有很好的连续性、抗噪性,同时对低质量指纹图像具有良好的鲁棒性,能更精确地反应指纹脊线的结构特征,而且计算量比现行方向图减少15%,试验证明了方法的有效性.