多偏好逻辑GMPL

针对缺乏多类型偏好共存的偏好逻辑系统的现状,MPL(logic of many kinds of preference)被构造为一种能够表示和推理四类型偏好的偏好逻辑,但是MPL的语义基于全前序偏好结构,因而不能表示不完全偏好.为此,提出了偏好逻辑GMPL(a generalized edition of MPL).此外,通过常见逻辑偏好的GMPL重写表明GMPL较强的表达能力和实际应用前景.并提出一种将GMPL的SAT问题归结为命题逻辑的SAT问题的方法.     

基于双极偏好占优的高维目标进化算法

高维目标优化是目前多目标优化领域的研究热点和难点.提出一种占优机制,即双极偏好占优用于处理高维目标优化问题.该占优机制同时考虑决策者的正偏好和负偏好信息,在非支配解之间建立了更加严格的占优关系,能够有效减少种群中非支配解的比例,引导算法向靠近正偏好同时远离负偏好的Pareto最优区域收敛.为检验该方法的有效性,将双极偏好占优融入NSGA-Ⅱ中,形成算法2p-NSGA-Ⅱ,并在2到15目标标准测试函数上进行测试,得到了良好的实验结果.同时,将所提出的占优机制与目前该领域的两种占优机制g占优和r占优进行性能对比,实验结果表明,2p-NSGA-Ⅱ算法无论是在求解精度还是运行效率上,整体上均优于g-NSGA-Ⅱ和r-NSGA-Ⅱ.     

正则化和交叉验证在组合预测模型中的应用

组合预测模型的权重确定方式对于提高模型精度至关重要,为研究正则化与交叉验证是否能改善组合预测模型的预测效果,提出将正则化和交叉验证应用于基于最小二乘法的组合预测模型.通过在组合模型的最优化求解中分别加入L₁、L₂范数正则化项,并对数据集进行留一交叉验证后发现：L₁、L₂范数正则化都对组合模型的预测精度具有改善效果,且L₁范数正则化比L₂范数正则化对组合预测模型的改善效果更好,并且参与组合预测的单项预测模型越多,正则化的改善效果越好,交叉验证对组合预测模型的改善效果则与给定实验数据量呈现正相关.     

γ演算到Action演算的转换

Action演算簇(action calculi)作为描述不同并发交互行为的数学框架,可以表示一大类具有某些相同特性的并发形式化模型.试图把(演算(一种基于约束的高阶并发计算模)也包含在action演算簇的框架下.首先定义了一个具体的action演算AC(K_γ),然后给出了从(演算到AC(K_γ)转换的形式描述,最后在定义AC(K_γ)的可观察性、弱互模拟关系和弱等价关系的基础上,以(演算为中间表示,证明了这种转换保持了(演算的弱行为等价性.研究表明,action演算簇可以表示基于约束的并发模型,从而充分说明了action演算簇的描述能力,并且为在action演算簇框架下把(演算与其他并发模型结合并进行比较提供了前提.     

描述逻辑εL混合循环术语集的LCS和MSC推理

分析了描述逻辑循环术语集的研究现状和存在的问题,在F.Baader工作的基础上进一步研究了描述逻辑εL混合循环术语集的LCS(least common subsumer)和MSC(most specific concept)推理问题.给出了εL混合循环术语集的语法和语义.针对εL混合循环术语集LCS和MSC推理的需要,提出了TBox-完全的概念,并重新定义了描述图.使用描述图和TBox-完全给出了最大不动点语义下εL混合循环术语集LCS和MSC的推理算法,证明了推理算法的正确性,并证明了推理算法是多项式时间复杂的.该推理算法为(L混合循环术语集的LCS和MSC推理提供了理论基础.     

不同通信模型下的全光树环网波长分配算法

研究了波分复用全光树环网在不同通信模型下的波长分配算法及其最坏性能分析.对于静态模型,证明了5L/2是树环网所需波长数的紧界.对于动态模型,提出了一种近似比为∑_i=1^hmax_r∈Ri[log|V(r)|]+h的波长分配算法,其中h为树环网的基树的层数,R_i为树环网中处于第i层的环的集合,|V(r)|为环r上的节点数.对于增量模型,提出了一种近似度为O[log₂(t+1)]的波长分配算法,其中t为树环网中的环数.     

公交网络下的一种费用限制最小时态路径查询索引

私人交通网络下的最短路径查询主要考虑路径长度、行驶时间等因素,而公共交通网络下的路径查询需要考虑路径上相邻的边的时间顺序约束以及路径的费用.研究了公共交通网络下3种查询：给定起点、终点、时间区间和费用上限,查找在时间区间内不超过费用上限的最早到达路径、最晚出发路径和最短耗时路径.首先给出一种Dijkstra变种算法Dijk-CCMTP,在此基础上给出3类查询的查询算法.然后提出一种高效的索引结构ACCTL（approximate cost constrained time labelling）.ACCTL采用Dijk-CCMTP对图中的每个顶点预先计算部分从该顶点出发的和到达该顶点的基本路径.对于任意从起点s到终点d的查询,可以采用类似数据库表连接的方式从ACCTL中连接从a出发的和到达d的路径生成近似解,避免遍历原图搜索路径.ACCTL建立索引的时间复杂度是O（|V|·△_max·|E|·（log|E|+△_max））,其中,|V|表示顶点数,|E|表示边数,△_max表示顶点的最大度数.实验验证ACCTL索引支持的查询速度比Dijkstra的变种算法的查询速度快2~3个数量级,并分析了影响建立索引时间和空间大小的因素.     

k-Median近似计算复杂度与局部搜索近似算法分析

k-Median问题的近似算法研究一直是计算机科学工作者关注的焦点,现有研究结果大多是关于欧式空间和Metric空间的,一般距离空间k-Median的结果多年来一直未见.考虑一般距离空间k-Median问题,设d_max/d_min表示k-Median实例中与客户点邻接的最长边长比最短边长的最大者.首先证明d_max/d_min≤ω+ε的k-Median问题不存在近似度小于1+ω-1/e的多项式时间近似算法,除非,由此推出Metric k-Median问题不可近似到1+2/e,除非NP(∈)DTME(N^{O(log logn)}).然后给出k-Median问题的一个局部搜索算法,分析表明,若有d_max/d_min≤ω,则算法的近似度为1+ω-1/2.该结果亦适用于Metric k-Median,ω≤5时,局部搜索算法求解Metric k-Median的近似度为3,好于现有结果3+2/P.通过计算机实验,进一步研究了k-Median局部搜索求解算法的实际计算效果和该算法的改进方法.     

P2-Packing问题参数算法的改进

P₂-Packing问题是一个典型的NP难问题.目前这个问题的最好结果是时间复杂度为O^*(2^5.301k)的参数算法,其核的大小为15k.通过对P₂-packing问题的结构作进一步分析,提出了改进的核心化算法,得到大小为7k的核,并在此基础上提出了一种时间复杂度为O^*(2^4.142k)的参数算法,大幅度改进了目前文献中的最好结果.     

Logical formalisms for representing bipolar preferences

Bipolar preferences distinguish between negative preferences inducing what is acceptable by complementation and positive preferences representing what is really satisfactory. This article provides a review of the main logics for preference representation. Representing preferences in a bipolar logical way has the advantage of enabling us to reason about them, while increasing their expressive power in a cognitively meaningful way. In the article, we first focus on the possibilistic logic setting and then discuss two other logics: qualitative choice logic and penalty logic. Finally, an application of bipolar preferences querying systems is outlined. © 2008 Wiley Periodicals, Inc.     

Decision Tradeoff Using Example-Critiquing and Constraint Programming

We consider constructive preference elicitation for decision aid systems in applications such as configuration or electronic catalogs. We are particularly interested in supporting decision tradeoff, where preferences are revised in response to the available outcomes. In several user-involved decision aid systems we designed in the past, we were able to observe three generic tradeoff strategies that people like to use. We show how a preference model based on soft constraints is well-suited for supporting these strategies. Such a framework provides an agile preference model particularly powerful for preference revision during tradeoff analysis. We further show how to integrate the constraint-based preference model with an interaction model called example-critiquing. We report on user studies which show that this model offers significant advantages over the commonly used ranked-list model, especially when the decision problem becomes complex.     

Modelling and learning user preferences over sets

Although there has been significant research on modelling and learning user preferences for various types of objects, there has been relatively little work on the problem of representing and learning preferences over sets of objects. We introduce a representation language, DD-PREF, that balances preferences for particular objects with preferences about the properties of the set. Specifically, we focus on the depth of objects (i.e. preferences for specific attribute values over others) and on the diversity of sets (i.e. preferences for broad vs. narrow distributions of attribute values). The DD-PREF framework is general and can incorporate additional object- and set-based preferences. We describe a greedy algorithm, DD-Select, for selecting satisfying sets from a collection of new objects, given a preference in this language. We show how preferences represented in DD-PREF can be learned from training data. Experimental results are given for three domains: a blocks world domain with several different task-based preferences, a real-world music playlist collection, and rover image data gathered in desert training exercises.     

Limitations of the Program Memory and the Expressive Power of Dynamic Logics

For a dynamic logic L we study dynamic logics L_n for which programs allowed in formulas cannot use more than n variables. We prove that there exists a structure A of a finite signature such that for a wide class of dynamic logics L and for every natural number n the logic L_n+1 is more expressive over A than L_n. This result is based on a construction of some canonical form for the formulas of L_n over a free one-generated groupoid.     

Timed Modal Logics for Real-Time Systems

In this paper, a timed modal logic L _c is presented for the specification and verification of real-time systems. Several important results for L _c are discussed. First we address the model checking problem and we show that it is an EXPTIME-complete problem. Secondly we consider expressiveness and we explain how to express strong timed bisimilarity and how to build characteristic formulas for timed automata. We also propose a compositional algorithm for L _c model checking. Finally we consider several control problems for which L _c can be used to check controllability.     

Meta-logic programming for a synonymy logic

In order to give a suitable framework for a synonymy-based logic programming, we argue about the possibility of reducing fuzzy logic programming to classical logical programming. More precisely, we show that given a fuzzy program in a language L, we can translate it into an equivalent classical program in a (meta-)language \textL_m {\text{L}}_{m} in which every predicate name in L becomes a constant in \textL_m . {\text{L}}_{m} . This enables us to admit in \textL_m {\text{L}}_{m} meta-relations among predicates and therefore, in particular, the synonymy.