A tutorial on branch and cut algorithms for the maximum stable set problem

This tutorial provides an overview of various characteristics of effective branch and cut type algorithms for the maximum stable set problem. We discuss several facet‐defining inequalities for the stable set polytope along with their separation routines. In particular, we review implementation tweaks for the separation routines and reference empirical studies, illustrating the performance of these cutting planes for benchmark graphs. In addition to the polyhedral study, we present basic preprocessing, discuss heuristic methods particularly suited within a branch and cut framework, and examine a branching rule.     

基于关系数据库的粗糙集约简改进算法

以粗糙集约简和关系数据库为研究对象,提出了一种带有数据预处理和新启发式信息值的粗糙集约简改进算法.通过应用了该改进算法的DBRuduct工具进行实验,实验数据表明决策表的对象个数和约简计算时间之间成近似线性的关系;以Rosetta中的遗传算法求得的约简作为实验参照,该算法不但可以在含有不一致数据的情况下获得正确的核属性,而且还使约简算法求得的约简更加趋向于最小约简.     

Distributed MPC of constrained linear systems with time-varying terminal sets

This paper proposes a decoupling strategy for the Distributed Model Predictive Control (DMPC) for a network of dynamically-coupled linear systems. Like most DMPC approaches, the proposed approach has a terminal set and uses a Lyapunov matrix for the terminal cost in the online optimization problem for each system. Unlike them, the terminal set changes at every time step and the Lyapunov matrix is not block-diagonal. These features result in a less conservative DMPC formulation. The proposed method is easy to implement when the network is strongly connected (or when a central collector is used). Otherwise, the computations of the terminal set require the online solutions of a series of linear programming problems but can be speeded up significantly by preprocessing. Numerical examples showing these results are provided.     

一种结合二元蚁群和粗糙集的连续属性离散化算法

离散化是一个重要的数据预处理过程,在规则提取、知识发现、分类等研究领域都有广泛的应用。提出一种结合二元蚁群和粗糙集的连续属性离散化算法。该算法在多维连续属性候选断点集空间上构建二元蚁群网络,通过粗糙集近似分类精度建立蚁群算法适宜度评价函数,寻找全局最优离散化断点集。通过UCI数据集验证算法的有效性,实验结果表明,该算法具有较好的离散化性能。     

基于运行时纹理合成的纹理画刷

纹理画刷是一种交互式纹理生成工具,能够在用户控制下生成所需要的纹理.本文提出了一种基于运行时纹理合成的纹理画刷实现算法.纹理合成采用逐块合成的方式,每次将整个样本图以选定的位移放置到合成图中,然后用Graph Cut来决定最终的输出区域.在预处理过程中计算得到两个相同样本图间的最佳切合位移集合,在合成过程中,贴块位移搜索范围限定在由此集合及用户控制所决定的一个很小的范围内,使合成速度达到实时.另外通过对"孤立块"采取"虚拟贴块法"选取贴块位移,较好地保持了画刷所生成纹理的一致性.实验结果表明该纹理画刷的纹理生成速度满足与用户交互的需求,且生成纹理的质量高.     

Convex hull made easy

Given a finite set S of n points and a point q in R^d, deciding if q ϵconv S is a matter of one linear program of size n, after a very simple linear-time preprocessing of S.     

一种新的基于DCT变换的线性判别分析

对于高维复杂模式识别问题,传统的线性判别分析通常首先采用PCA变换来降低模式的维数,然后再求取最优判别矢量集。然而PCA变换是以判别信息的损失为代价的,故无法保证所提取的特征是最优的。DCT变换具有"能量聚集特性"和变换的保距特性,文中正是基于此特性,提出一种新的基于DCT变换的线性判别分析方法,同时,也给出了一种在该模型下的最优判别矢量集的直接求解方法。实验表明,文中算法具有计算速度快、识别率高的优点。     

粗集理论对股票时间序列的知识发现

提出了将粗集理论应用于时间序列的知识发现。知识发现的过程包括时间序列数据预处理、属性约简和规则抽取三部分。其中数据预处理主要用信号处理技术清洗数据,然后将清洗后的时间序列按照某个变量的变化趋势进行分割,分割后每个时间段内的变化趋势不变,从而将时间序列转换成为一系列静态模式(每种模式代表一种行为趋势),从而去掉其时间依赖性。把决定各种模式的相关属性抽取出来组成一个适用于粗集理论的信息表,然后采用粗集理论对信息表进行属性约简和规则抽取,所得到的规则可以用于预测时间序列在未来的行为。最后将该方法用于股票的趋势预测,取得良好效果。     

掌纹图像的获取及其预处理

提出了一种无限制的获取掌纹图像的方法,并使用该方法构建了掌纹库.该掌纹库可以成为掌纹识别算法训练集和测试集的来源,也可以成为进一步推进掌纹识别研究与发展的基础.在该掌纹库的基础上,对掌纹库中的掌纹图像进行了预处理,将原始图像二值化后利用定位点自动检测技术检测出掌纹图像中两个关键的定位点,并以此为基础对掌纹图像进行旋转校正,最后切取一定区域的掌纹子图,为进一步提取掌纹特征打下了较好的基础.     

高性能混凝土强度预测的神经网络-主成分分析

在误差逆传播算法神经网络预测模型数据前处理中,对样本集优化,采用多元统计分析中的主成分分析法,提取影响粉煤灰高性能混凝土抗压强度的主要因素,消除影响因素间的线性相关性。研究结果表明,用该方法处理后的样本数据输入神经网络,提高了预测效率,训练时间减少,预测精度也有一定程度的提高,网络结构得到简化。     

Static and dynamic preprocessing methods in neural networks

Preprocessing is recognized as an important tool in modeling, particularly when the data or underlying physical process involves complex nonlinear dynamical interactions. This paper will give a review of preprocessing methods used in linear and nonlinear models. The problem of static preprocessing will be considered first, where no dependence on time between the input vectors is assumed. Then, dynamic preprocessing methods which involve the modification of time-dependent input values before they are used in the linear or nonlinear models will be considered. Furthermore, the problem of an insufficient number of input vectors is considered. It is shown that one way in which this problem can be overcome is by expanding the weight vector in terms of the available input vectors. Finally, a new problem which involves both cases of: (1) transformation of input vectors; and (2) insufficient number of input vectors is considered. It is shown how a combination of the techniques used to solve the individual problems can be combined to solve this composite problem. Some open issues in this type of preprocessing methods are discussed.     

Quarterly Time-Series Forecasting With Neural Networks

Forecasting of time series that have seasonal and other variations remains an important problem for forecasters. This paper presents a neural network (NN) approach to forecasting quarterly time series. With a large data set of 756 quarterly time series from the M3 forecasting competition, we conduct a comprehensive investigation of the effectiveness of several data preprocessing and modeling approaches. We consider two data preprocessing methods and 48 NN models with different possible combinations of lagged observations, seasonal dummy variables, trigonometric variables, and time index as inputs to the NN. Both parametric and nonparametric statistical analyses are performed to identify the best models under different circumstances and categorize similar models. Results indicate that simpler models, in general, outperform more complex models. In addition, data preprocessing especially with deseasonalization and detrending is very helpful in improving NN performance. Practical guidelines are also provided.     

动态故障树的不交化定量分析方法

动态故障树被广泛应用于动态系统的可靠性分析中,其中割序描述了系统的失效模式,割序集的不交化可以简化顶点失效概率的求解,目前还没有有效的适用于动态故障树的不交化定量分析方法.提出了一种不交化定量分析方法:在割序的基础上融入时序逻辑提出扩展割序的概念,其与以往类似概念相比增强了表达能力;根据基事集和时限集分解最小扩展割序集,将其转换成不交化扩展割序集;再将不交化扩展割序转换成标准扩展割序,然后对其各割项进行冲突检测、时限集精简、基事集拓扑排序,以对标准扩展割序进行量化计算;并对该方法所涉及到的算法进行了详细的证明和时间复杂性分析.最后将其应用到一个案例中,并同基于inclusion-exclusion规则的MCS方法进行了对比,实验结果显示该方法的时间开销明显降低.该方法可以获得动态故障树的不交化扩展割序集,降低求解时间开销.     

在动态树中寻找祖先

文章提出了有线性预处理时间,在恒定的查询时间内,通过加叶和加根操作对树的祖先进行查找的一种有效算法,这种算法能在O(m+n)时间内进行m次祖先查询及n次增叶操作。     

有效预处理P2P 网络中的子空间skyline 查询

多维空间的skyline 查询处理是近年来数据库领域的一个研究重点和热点.Vlachou 等人首次考虑如何在P2P 网络中有效进行子空间上的skyline 查询,并提出“扩展skyline 集合”的概念来减少预处理时的网络传输量.然而实验评估表明,扩展skyline 集合只能有限地减少子空间skyline 查询预处理的数据传输量.基于此,提出一种缩减处理时数据传输量的有效方法TPAOSS(three-phase algorithm for optimizing skyline scalar).TPAOSS 算法根据全空间skyline 集合与子空间skyline 集合间的语义关系分3 个阶段来传输必要的数据,其中第1 阶段发送全空间skyline 对象;第2 阶段接收种子skyline 对象;而第3 阶段基于Bloom filter 技术发送种子skyline 对象在子空间上的重复对象.为了降低第2 阶段的数据传输量,给出两种接收种子skyline 对象的有效策略.理论分析和实验评估结果表明,所给出的算法具有有效性和实用性.     

A Linear Algorithm for Collision Detection of 3-Dimensional Convex Polytopes

Given two disjoint 3-dimensional convex polytopes P and Q and a straight direction along Which P moves in translation, this paper presents a linear algorithm for determining Whether P collides with Q, and the possible collision positions on P and Q. This result is achieved by using the hierarchicat representation of polytopes, of which the preprocessing time is linear with space.     

基于目标演绎距离的一阶逻辑子句集预处理方法

一阶逻辑定理证明是人工智能的核心基础,研究一阶逻辑自动定理证明器的相关理论和高效的算法实现具有重要的学术意义。当前一阶逻辑自动定理证明器首先通过子句集预处理约简子句集规模,然后通过演绎方法对定理进行判定。现有的应用于证明器中的子句集预处理方法普遍只从与目标子句项符号相关性角度出发,不能很好地从文字的互补对关系中体现子句间的演绎。为了在子句集预处理时从演绎的角度刻画子句间的关系,定义了目标演绎距离的概念并给出了计算方法,提出了一种基于目标演绎距离的一阶逻辑子句集预处理方法。首先对原始子句集进行包含冗余子句约简并应用纯文字删除规则,然后根据目标子句计算剩余子句集中的文字目标演绎距离、子句目标演绎距离,并最终通过设定子句演绎距离阈值来实现对子句集的进一步预处理。将该预处理方法应用于顶尖证明器Vampire,以2017年国际一阶逻辑自动定理证明器标准一阶逻辑问题组竞赛例为测试对象,在标准的300 s内,加入提出的子句集预处理方法的Vampire4.1相比原始的Vampire4.1多证明4个定理,能证明10个Vampire4.1未证明的定理,占其未证明定理总数的13.5%;在证明的定理中,提出的子句集预处理方法能对77.2%的子句集进行约简,最大子句集约简规模达到51.7%。实验结果表明,提出的一阶逻辑子句集预处理方法是一种有效的方法,能有效地约简一阶逻辑子句集的规模,提高一阶逻辑自动定理证明器的证明能力。     

Travel Plans in Public Transit Networks Using Artificial Intelligence Planning Models

Users of public transit networks require tools that generate travel plans to traverse them. The main issue is that public transit networks are time and space dependent. Travel plans depend on the current location of users and transit units, along with a set of user preferences and time restrictions. In this work, we propose the design and development of artificial intelligence (AI) planning models for engineering travel plans for such networks. The proposed models consider temporal actions, bus locations, and user preferences as constraints, to restrict the set of travel plans generated. Our approach decouples model design from algorithm construction, providing a greater level of flexibility and richness of solutions. We also introduce an integer linear programming formulation, and a fast preprocessing procedure, to evaluate the quality of the solutions returned by the proposed planning models. Experimental results show that AI planning models can efficiently generate close to optimal solutions. Furthermore, our analysis identifies user preferences as the most critical factor that increases solution complexity for planning models.     

View Planning for Multi-View Stereo 3D Reconstruction Using an Autonomous Multicopter

Multi-view stereo algorithms are an attractive technique for the digital reconstruction of outdoor sites. Concerning the data acquisition process a vertical take off and landing UAV carrying a digital camera is a suitable platform in terms of mobility and flexibility in viewpoint placement. We introduce an automated UAV based data acquisition and outdoor site reconstruction system. A special focus is set on the problem of model based view planning using a coarse digital surface model (DSM) with minimal data preprocessing. The developed view planning heuristic considers a coverage, a maximum view angle and an overlapping constraint imposed by multi-view stereo reconstruction techniques. The time complexity of the algorithm is linear with respect to the size of the area of interest. We demonstrate the efficiency of the entire system in two scenarios, a building and a hillside.     

基于压缩全文索引的演变图查询

演变图中含有大量的时间和空间信息,其中某些空间信息随着时间的推移表现出相似的演变规律。给出了一种演变图查询模型,可以挖掘出在相同时间范围内具有相同变化规律的演变子图。但是演变图的规模往往是巨大的,当需要对其进行多次查询时,每次遍历整个演变图将带来非常高的查询代价,而现有的基于枚举的哈希索引算法又使得预处理过程拥有相当大的时间和空间开销,为了减少对大规模演变图的预处理代价,将压缩的全文索引技术应用于演变图,它基于涡轮转换和后缀数组。在构建后缀数组时,给出了两种不同的线性算法,确保了预处理过程的稳定性。通过在Facebook、Enron邮件系统以及模拟数据集上的实验,评估了该算法的可行性、效率以及可扩展性。