独立多处理机任务静态调度问题的近似算法

研究独立多处理机任务静态调度问题P_m|fix|C_max,即在m个处理机系统中调度n个多处理机任务,每个任务指派到所需一组处理机上不可剥夺地执行.该问题应用广泛但早已证明为NP难问题,而且也不存在常数近似算法.分析了问题P_m|fix|C_max和其中所有任务都是单位处理机时间的特殊情形P_m|fix,p=1|C_max的调度,并利用实例划分(split scheduling,简称SS)、首次满足优先(first fit,简称FF)和最大宽度优先(large wide first,简称LWF)等方法,构造了问题P_m|fix,p=1|C_max的√2m +1近似算法和问题P_m|fix|C_max的2√m 近似算法,优于目前已有文献的最好结果.     

关于有ω-穷自动机的两个新的接受条件*

至今被公开的ω-有穷自动机的接受条件有6个即C₁—C₆，寻找新的接受条件和研究ω-有穷自动机关于新接受条件接受ω-语言的能力是ω-有穷自动机理论中的一个重要课题．本文定义了ω-有穷自动机的两个新的接受条件Z₁和Z₂，并且研究了：(1)ω-U—NFA关于Z_i(i=1,2)接受 ω-语言的能力，得到了 N     

Grzegorczyk分层的一种延伸

本文讨论某些递归函数类的分层问题．首先给出的是原始的Ｇｏｒｚｅｇｏｒｃｚｙｋ分层的一种较为简单的等价定义．然后，作为对Ａｃｋｅｒｍａｎｎ函数的一种推广，定义了一个递归函数序列｛Ａ_ｎ｝_ｎ∈ω．并以此作为分层函数列定义了一种新的递归分层｛Ｚ_ｎ｝_ｎ∈ω（即Ｚ—分层），这种分层涉及了比原始递归函数类更大的一个违归函数类．实际上，原始递归函数类仅是Ｚ—分层的第一层Ｚ_０．而且这种分层的任意的第ｎ＋１层都含     

数据仓库系统中层次式Cube存储结构

区域查询是数据仓库上支持联机分析处理(on-line analytical processing,简称OLAP)的重要操作.近几年,人们提出了一些支持区域查询和数据更新的Cube存储结构.然而这些存储结构的空间复杂性和时间复杂性都很高,难以在实际中使用.为此,提出了一种层次式Cube存储结构HDC(hierarchical data cube)及其上的相关算法.HDC上区域查询的代价和数据更新代价均为O(log^dn),综合性能为O((logn)^2d)(使用C_qC_u模型)或O(K(logn)^d)(使用C_qn_q+C_un_u模型).理论分析与实验表明,HDC的区域查询代价、数据更新代价、空间代价以及综合性能都优于目前所有的Cube存储结构.     

有中断时间代价的一致并行机抢先调度问题

提出了一种具有中断时间代价的抢先调度问题(P|ptmn(δ)|C_max):在抢先调度中,一个任务发生一次中断,其总的执行时间会增加一个δ.该问题在工程任务分配、分布式计算和网络通信等实际问题中有着广泛的应用背景.证明了这是一个NP-hard问题,给出了一个时间复杂度为O(nlogn+m)的脱线近似算法LPT-Wrap,其近似比小于等于1.40825,并分析了P|ptmn(δ)|C_max的在线特性,给出一个线性时间复杂度的在线近似算法,其竞争比为2.     

关于二元延迟3步前馈逆有限自动机的结构

前馈逆有限自动机的结构是有限自动机可逆性理论中的基本问题.对延迟步数≥3的前馈逆结构的刻划,则是一个长期的未解决问题.研究了二元延迟3步前馈逆有限自动机的结构.对于自治有限自动机M_a的状态图为圈的二元延迟3步弱可逆半输入存储有限自动机C(M_af ),给出了其长3极小输出权分别为1,2,8三种情形下结构的一种刻画.由于C(M_af )延迟3步弱可逆当且仅当它是延迟3步弱逆,因此,得到了二元延迟3步前馈逆有限自动机结构的一种部分刻画.     

线性动态系统反演模型在人体行为识别的应用

目的 人体行为识别在视频监控、环境辅助生活、人机交互和智能驾驶等领域展现出了极其广泛的应用前景。由于目标物体遮挡、视频背景阴影、光照变化、视角变化、多尺度变化、人的衣服和外观变化等问题,使得对视频的处理与分析变得非常困难。为此,本文利用时间序列正反演构造基于张量的线性动态模型,估计模型的参数作为动作序列描述符,构造更加完备的观测矩阵。方法 首先从深度图像提取人体关节点,建立张量形式的人体骨骼正反向序列。然后利用基于张量的线性动态系统和Tucker分解学习参数元组（A_F,A_I,C）,其中C表示人体骨架信息的空间信息,A_F和A_I分别描述正向和反向时间序列的动态性。通过参数元组构造观测矩阵,一个动作就可以表示为观测矩阵的子空间,对应着格拉斯曼流形上的一点。最后通过在格拉斯曼流形上进行字典学习和稀疏编码完成动作识别。结果 实验结果表明,在MSR-Action 3D数据集上,该算法比Eigenjoints算法高13.55%,比局部切从支持向量机（LTBSVM）算法高2.79%,比基于张量的线性动态系统（tLDS）算法高1%。在UT-Kinect数据集上,该算法的行为识别率比LTBSVM算法高5.8%,比tLDS算法高1.3%。结论 通过大量实验评估,验证了基于时间序列正反演构造出来的tLDS模型很好地解决了上述问题,提高了人体动作识别率。     

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.     

三机冗余容错系统的描述和验证

使用XYZ/E描述和验证三机冗余容错系统.考虑每台计算机加载了一个不断向外界环境输出数据的确定性顺序程序P,用XYZ/E程序SingleProcessor_P刻画程序P在单机上运行,用时序逻辑式Spec_P刻画P向外部环境输出的数据所满足的性质.最后证明,采用三机冗余模式所得到的程序TripleProcessors_P即使在出现硬件错误的情况下运行,也能满足性质Spec_P.     

嵌套谓词等式系与弱互模拟

带赋值符号迁移图是一般传值进程的语义模型,其强互模拟等价可以归结为谓词等式系的最大解.该文将这一结果推广到弱互模拟等价,为此,引入嵌套谓词等式系的概念,并提出算法,将带赋值符号迁移图的弱互模拟等价归结为形如E₂μE₁的嵌套谓词等式系的最大解.     

THE CATEGORY THEORY OF TIME SYSTEMS

In this paper we show a categorical treatment of general time systems using the categorization method presented in our previous paper. Various concepts about general time systems are categorized in the unified framework. Some category theoretical tools for the investigation of such time systems are presented. Using those tools some basic properties of time systems are explored in our framework. In particular, a conceptual equivalence between the causality and the state space representability is proved in the categorical terms. These results show that our method can be a universal tool for a categorization and a categorical treatment of mathematically defined general systems.     

Pattern recognition categories

We present a categorical definition of pattern recognition, which unifies algebraic and topological formalisms. We state and prove a recognition theorem, and show how to define equivalence and invariants for both settings.     

A calculus of coroutines

We describe a simple but expressive calculus of sequential processes, represented as coroutines. We show that this calculus can be used to express a variety of programming language features including procedure calls, labelled jumps, integer references and stacks. We describe the operational properties of the calculus using reduction rules and equational axioms.

We describe a notion of categorical model for our calculus, and give a simple example of such a model based on a category of games and strategies. We prove full abstraction results showing that equivalence in the categorical model corresponds to observational equivalence in the calculus, and also to equivalence of evaluation trees, which are infinitary normal forms for the calculus.

We show that our categorical model can be used to interpret the untyped λ-calculus and use this fact to extract a sound translation of the latter into our calculus of coroutines.     

The logic of message-passing

Message-passing is a key ingredient of concurrent programming. The purpose of this paper is to describe the equivalence between the proof theory, the categorical semantics, and term calculus of message-passing. In order to achieve this we introduce the categorical notion of a linear actegory and the related polycategorical notion of a poly-actegory. Not surprisingly the notation used for the term calculus borrows heavily from the (synchronous) π-calculus. The cut-elimination procedure for the system provides an operational semantics.     

On the efficiency of categorical combinators as a rewriting system

Categorical combinators form a formal system similar to Curry's combinatory logic. The original system was developed by Curien, inspired by the equivalence of the theories of typed λ-calculus and Cartesian closed categories, as shown by Lambek and Scott. A new system for categorical combinators was introduced by the author. This system uses a more compact notation for the code and needs a smaller set of rewriting rules. The aim of this paper is to analyse these two different rewriting systems for categorical combinators as a basis for implementation of applicative languages, and compare them with the classical approach due to Turner, using combinatory logic.     

A congruence rule format for name-passing process calculi

We introduce a GSOS-like rule format for name-passing process calculi. Specifications in this format correspond to theories in nominal logic. The intended models of such specifications arise by initiality from a general categorical model theory. For operational semantics given in this rule format, a natural behavioural equivalence—a form of open bisimilarity—is a congruence.     

Statistical inference for equivalence trials with ordinal responses: A latent normal distribution approach

Testing of equivalence/non-inferiority has become an essential component in modern drug and treatment assessment. Before a newly developed treatment is introduced and applied to its target population, it is necessary to compare it to an existing (reference/standard) treatment. Unlike the traditional trial of testing the equality between two treatments, an equivalence trial, for instance, attempts to demonstrate that the responses to two treatments differ by an amount which is clinically insignificant. In many applications, the outcome measures of interest are usually recorded in ordinal scale (e.g., very good; good; moderate; poor). This paper presents a simple approach to the problem of equivalence testing in the presence of ordered categorical data. The proposed methodology operates on the assumption that the observed ordinal variable is governed by an underlying normally distributed trait. The new approach can be readily adopted for (i) commonly used equivalence limits such as difference and the ratio of treatment means and (ii) both one-sided non-inferiority and two-sided equivalence trials. We illustrate our methodology with two medical examples and demonstrate how test results and confidence interval estimates can be obtained from a freely available computer program.     

A Categorical Critical-pair Completion Algorithm

We introduce a general critical-pair/completion algorithm, formulated in the language of category theory. It encompasses the Knuth–Bendix procedure for term rewriting systems (also modulo equivalence relations), the Gröbner basis algorithm for polynomial ideal theory, and the resolution procedure for automated theorem proving. We show how these three procedures fit in the general algorithm, and how our approach relates to other categorical modeling approaches to these algorithms, especially term rewriting.     

Domain semantics of possibility computations

The purpose of this paper is to introduce a kind of computational model, called possibility computation, to deal with non-determinism. Both its denotational and logical semantics in the framework of domain theory are established and their equivalence is verified. In this approach, the denotational semantics will be set up by assigning to programs possibility state transformers, i.e., Scott-continuous functions from the domain of input states to the possibility powerdomain of the codomain of final states. This possibility powerdomain of a dcpo will consist of all possibility valuations, ordered pointwise, of this dcpo. The logical semantics will be given by fuzzy predicate transformers introduced by the first author and Jung following Dijkstra’s predicate transformers. Two semantics equivalence will be verified in terms of the integration of fuzzy predicates with respect to possibility valuations. The categorical monad of the possibility powerdomain will be investigated.     

一种基于多属性权重的分类数据子空间聚类算法

采用多属性频率权重以及多目标簇集质量聚类准则,提出一种分类数据子空间聚类算法.该算法利用粗糙集理论中的等价类,定义了一种多属性权重计算方法,有效地提高了属性的聚类区分能力;在多目标簇集质量函数的基础上,采用层次凝聚策略,迭代合并子簇,有效地度量了各类尺度的聚类簇;利用区间离散度,解决了使用阈值删除噪音点所带来的参数问题;利用属性对簇的依附程度,确定了聚类簇的属性相关子空间,提高了聚类簇的可理解性.最后,采用人工合成、UCI和恒星光谱数据集,实验验证了该聚类算法的可行性和有效性.