引入线索约束的设计模式变体挖掘研究

针对设计模式变体挖掘准确率较低的问题,提出引入线索约束的设计模式变体挖掘方法,旨在基于约束满足问题CSP描述设计模式变体线索,给出DPVMC算法,分结构特征约束与时序特征约束2个阶段引入线索.以Proxy、Bridge、Command、Factory Method模式变体为例,设计了2阶段的单个设计模式变体挖掘实验与集...     

Relatively quantified constraint satisfaction

The constraint satisfaction problem (CSP) is a convenient framework for modelling search problems; the CSP involves deciding, given a set of constraints on variables, whether or not there is an assignment to the variables satisfying all of the constraints. This paper is concerned with the more general framework of quantified constraint satisfaction, in which variables can be quantified both universally and existentially. We study the relatively quantified constraint satisfaction problem (RQCSP), in which the values for each individual variable can be arbitrarily restricted. We give a complete complexity classification of the cases of the RQCSP where the types of constraints that may appear are specified by a constraint language.     

Maximum H-colourable subdigraphs and constraint optimization with arbitrary weights

In the maximum constraint satisfaction problem (Max CSP), one is given a finite collection of positive-weight constraints on overlapping sets of variables, and the goal is to assign values from a given domain to the variables so that the total weight of satisfied constraints is maximized. We consider this problem and its variant Max AW CSP where the weights are allowed to be both positive and negative, and study how the complexity of the problems depends on the allowed constraint types. We prove that Max AW CSP over an arbitrary finite domain exhibits a dichotomy: it is either polynomial-time solvable or NP-hard. Our proof builds on two results that may be of independent interest: one is that the problem of finding a maximum H-colourable subdigraph in a given digraph is either NP-hard or trivial depending on H, and the other a dichotomy result for Max CSP with a single allowed constraint type.     

A note on some collapse results of valued constraints

Valued constraint satisfaction problem (VCSP) is an optimisation framework originally coming from Artificial Intelligence and generalising the classical constraint satisfaction problem (CSP). The VCSP is powerful enough to describe many important classes of problems. In order to investigate the complexity and expressive power of valued constraints, a number of algebraic tools have been developed in the literature. In this note we present alternative proofs of some known results without using the algebraic approach, but by representing valued constraints explicitly by combinations of other valued constraints.     

A CSP approach for the network of product lifecycle constraints consistency in a collaborative design context

In this paper, we are considering that the design process can be modelled in the form of a constraint satisfaction problem (CSP). CSP modelling or resolution has proved its efficiency within the framework of single-designer design. We propose to extend the functions of CSP to the context of multi-concept design of the same artefact. We define CoCSP as cooperative constraint satisfaction problem including the actors of the design problem. We are presenting the operating principles of an algorithm for the real-time management of design decisions, based on a model described in the form of a CoCSP for the integration of supply-chain constraints. This algorithm enables the number of design decisions rejected at a given moment in design to be kept to a minimum. The algorithm forms the core of a prototype for an unsupervised, generic constraint-based collaborative design system. Our aim is to produce a platform centred on the notion of constraints that will enable a product design problem to be modelled and solved by integrating supply-chain constraints as far upstream as possible.     

Structural decompositions for problems with global constraints

A wide range of problems can be modelled as constraint satisfaction problems (CSPs), that is, a set of constraints that must be satisfied simultaneously. Constraints can either be represented extensionally, by explicitly listing allowed combinations of values, or implicitly, by special-purpose algorithms provided by a solver. Such implicitly represented constraints, known as global constraints, are widely used; indeed, they are one of the key reasons for the success of constraint programming in solving real-world problems. In recent years, a variety of restrictions on the structure of CSP instances have been shown to yield tractable classes of CSPs. However, most such restrictions fail to guarantee tractability for CSPs with global constraints. We therefore study the applicability of structural restrictions to instances with such constraints. We show that when the number of solutions to a CSP instance is bounded in key parts of the problem, structural restrictions can be used to derive new tractable classes. Furthermore, we show that this result extends to combinations of instances drawn from known tractable classes, as well as to CSP instances where constraints assign costs to satisfying assignments.     

Consistency techniques for continuous constraints

We consider constraint satisfaction problems with variables in continuous, numerical domains. Contrary to most existing techniques, which focus on computing one single optimal solution, we address the problem of computing a compact representation of the space of all solutions admitted by the constraints. In particular, we show how globally consistent (also called decomposable) labelings of a constraint satisfaction problem can be computed.Our approach is based on approximating regions of feasible solutions by 2 ^k -trees, a representation commonly used in computer vision and image processing. We give simple and stable algorithms for computing labelings with arbitrary degrees of consistency. The algorithms can process constraints and solution spaces of arbitrary complexity, but with a fixed maximal resolution.Previous work has shown that when constraints are convex and binary, path-consistency is sufficient to ensure global consistency. We show that for continuous domains, this result can be generalized to ternary and in fact arbitrary n-ary constraints using the concept of (3,2)-relational consistency. This leads to polynomial-time algorithms for computing globally consistent labelings for a large class of constraint satisfaction problems with continuous variables.     

Automatic Generation of Redundant Models for Permutation Constraint Satisfaction Problems

If we have two representations of a problem as constraint satisfaction problem (CSP) models, it has been shown that combining the models using channeling constraints can increase constraint propagation in tree search CSP solvers. Handcrafting two CSP models for a problem, however, is often time-consuming. In this paper, we propose model induction, a process which generates a second CSP model from an existing model using channeling constraints, and study its theoretical properties. The generated induced model is in a different viewpoint, i.e., set of variables. It is mutually redundant to and can be combined with the input model, so that the combined model contains more redundant information, which is useful to increase constraint propagation. We also propose two methods of combining CSP models, namely model intersection and model channeling. The two methods allow combining two mutually redundant models in the same and different viewpoints respectively. We exploit the applications of model induction, intersection, and channeling and identify three new classes of combined models, which contain different amounts of redundant information. We construct combined models of permutation CSPs and show in extensive benchmark results that the combined models are more robust and efficient to solve than the single models.     

Shape optimization of heavy load carrying components for structural performance and manufacturing cost

This paper addresses the trade-off between structural performance and manufacturing cost of heavy load carrying components by incorporating virtual machining (VM) technique in computer-aided design (CAD)-based shape optimization problem. A structural shape optimization problem is set up to minimize total cost, subject to the limits on structural performance measures. For every design iteration, finite element analysis (FEA) is conducted to evaluate structural performance, and VM is employed to ascertain machinability and estimate machining time. Design sensitivity coefficients of objective function and constraints are computed and supplied to the optimization algorithm. Based on the gradients, the algorithm determines design changes, which are used to update FEA and VM models. The process is repeated until specified convergence criterion is satisfied. Application programs developed to integrate commercially available CAD/CAM/FEA/Design optimization tools enable implementation in virtual environment and facilitate automation. The application programs can be reused for similar design problems provided that the same set of tools is used.     

Periodic Constraint Satisfaction Problems: Tractable Subclasses

We study a generalization of the constraint satisfaction problem (CSP), the periodic constraint satisfaction problem. An input instance of the periodic CSP is a finite set of generating constraints over a structured variable set that implicitly specifies a larger, possibly infinite set of constraints; the problem is to decide whether or not the larger set of constraints has a satisfying assignment. This model is natural for studying constraint networks consisting of constraints obeying a high degree of regularity or symmetry. Our main contribution is the identification of two broad polynomial-time tractable subclasses of the periodic CSP.     

Neural networks for finite constraint satisfaction

Constraint Satisfaction Problems (CSPs) are in general NP-hard, and a general deterministic polynomial time algorithm is not known. They play a central role in real-life problems. The satisfaction of a Conjunctive Normal Form (CNF-SAT)is the core of any CSP. We present a new modelisation technique for any CSP with finite variable domains, and, in particular, for solving CNF-SAT. The knowledge representation is based on two fundamental types of constraint: the choice constraint, and the exclusion constraint. These models are then implemented by means of several different neural networks, some based on backpropagation learning and others on different procedures. All these networks are trained through a supervised procedure, and learn to efficiently solve CNF-SAT. The results of significant tests are described: they show that some networks can effectively solve the proposed problems.     

The distributed constraint satisfaction problem: formalization andalgorithms

We develop a formalism called a distributed constraint satisfaction problem (distributed CSP) and algorithms for solving distributed CSPs. A distributed CSP is a constraint satisfaction problem in which variables and constraints are distributed among multiple agents. Various application problems in distributed artificial intelligence can be formalized as distributed CSPs. We present our newly developed technique called asynchronous backtracking that allows agents to act asynchronously and concurrently without any global control, while guaranteeing the completeness of the algorithm. Furthermore, we describe how the asynchronous backtracking algorithm can be modified into a more efficient algorithm called an asynchronous weak-commitment search, which can revise a bad decision without exhaustive search by changing the priority order of agents dynamically. The experimental results on various example problems show that the asynchronous weak-commitment search algorithm is, by far more, efficient than the asynchronous backtracking algorithm and can solve fairly large-scale problems     

随机约束满足问题的回溯算法分析

提出一种新的随机CSP(constraint sa tisfaction problem)模型,并且通过研究搜索树的平均节点数,分析了回溯算法求解该模型 的平均复杂性.结果表明,这种模型能够生成难解的CSP实例,找到所有的解或证明无解所需的 平均节点数即随变量数的增加而指数增长.因此,该模型可以用来研究难解实例的性质和CSP 算法的性能等问题,从而有助于设计出更为高效的算法.     

Alliance of model-driven engineering with a proof-based formal approach

Model-driven engineering (MDE) promotes the use of models throughout the software development cycle in order to increase abstraction and reduce software complexity. It favors the definition of domain-specific modeling languages (DSMLs) thanks to frameworks dedicated to meta-modeling and code generation like EMF (Eclipse Modeling Framework). The standard semantics of meta-models allows interoperability between tools such as language analysers (e.g., XText), code generators (e.g., Acceleo), and also model transformation tools (e.g., ATL). However, a major limitation of MDE is the lack of formal reasoning tools allowing to ensure the correctness of models. Indeed, most of the verification activities offered by MDE tools are based on the verification of OCL constraints on instances of meta-models. However, these constraints mainly deal with structural properties of the model and often miss out its behavioral semantics. In this work, we propose to bridge the gap between MDE and the rigorous world of formal methods in order to guarantee the correctness of both structural and behavioral properties of the model. Our approach translates EMF meta-models into an equivalent formal B specification and then injects models into this specification. The equivalence between the resulting B specification and the original EMF model is kept by proven design steps leading to a rigorous MDE technique. The AtelierB prover is used to guarantee the correctness of the model’s behavior with respect to its invariant properties, and the ProB model-checker is used to animate underlying execution scenarios which are translated back to the initial EMF model. Besides the use of these automatic reasoning tools in MDE, proved B refinements are also investigated in this paper in order to gradually translate abstract EMF models to concrete models which can then be automatically compiled into a programming language.

     

Designer’s preferences regarding equality constraints in robust design optimization

Robust design optimization (RDO) problems can generally be formulated by appropriately incorporating uncertainty into the corresponding deterministic optimization problems. Equality constraints in the deterministic problem need to be carefully formulated into the RDO problem because of the strictness associated with their feasibility. In this context, equality constraints have been generally classified into two types: (1) those that must be satisfied regardless of uncertainty, examples include physics-based constraints, such as F = ma, and (2) those that cannot be satisfied because of uncertainty, which are typically designer-imposed, such as dimensional constraints. This paper addresses the notion of preferred degree of satisfaction of deterministic equality constraints under uncertainty. Whether or not a particular equality constraint can be exactly satisfied depends on the statistical nature of the design variables that exist in the constraint and on the designer’s preferences. In this context, this paper puts forth three contributions. First, we develop a comprehensive classification of equality constraints in a way that is mutually exclusive and collectively exhaustive. Second, we present a rank-based matrix approach to interactively classify equality constraints, which systematically incorporates the designer’s preferences into the classification process. Third, we present an approach to incorporate the designer’s intra-constraint and inter-constraint preferences for designer-imposed constraints into the RDO formulation. Intra-constraint preference expresses how closely a designer wishes to satisfy a particular constraint; for example, in terms of its mean and standard deviation. A designer may express inter-constraint preference if satisfaction of a particular designer-imposed constraint is more important than that of another. We present an optimization formulation that incorporates the above discussed constraint preferences, which provides the designer with the means to explore design space possibilities. The formulation entails interesting implications in terms of decision making. Two engineering examples are provided to illustrate the practical usefulness of the developments proposed in this paper.     

Tractable classes for CSPs of arbitrary arity: from theory to practice

The research of this thesis focuses on the analysis of polynomial classes and their practical exploitation for solving constraint satisfaction problems (CSPs) with finite domains. In particular, I worked on bridging the gap between theoretical works and practical results in constraint solvers. Specifically, the goal of this thesis is to find explanation for the effectiveness of solvers, and also to show that studied tractable classes are not artificial since several real-problems among the ones used in the CSP 2008 Competition belong to them.Our work is organized into three main parts. In the first part, we proposed several types of microstructures for CSPs of arbitrary arity which are based on some knwon binary encoding of non-binary CSPs like, dual encoding, hidden-variable transformation and mixed (or double) encoding. These theoretical tools are designed to facilitate the study of tractable classes, sets of CSP instances which can be solved in polytime, when the constraints are non-binary. After that, we propose a new tractable classes of CSPs whose the highlighting should allow on the one hand to explain the effectiveness of solvers of the state of the art namely FC, MAC, RFL and on the second hand to provide the opportunities for easy integration in these solvers. These would include the definition of new tractable classes without using of an ad hoc algorithms as in the traditional case. These new tractable classes are related to the number of maximal cliques in the microstructure of binary or non-binary CSP. In the last part, we focus on the presence of instances belonging to polynomial classes in classical benchmarks used by the CP community. We study in particular the Broken-Triangle Property (BTP) and its extension DBTP to CSP of arbitrary arity. Next, we prove that BTP can also be used to reduce the size of the search space by merging pairs of values on which no broken triangle exists. Finally, we introduce a formal framework, called transformation, and we develop the concept of hidden tractable class that we exploit from an experimental point of view.     

CSPs with counters: a likelihood-based heuristic

Abstract. Counter constraints are a naturalrepresentation of constraints on the finite capacity of resources in resource-allocation type problems. They are a generic family of non-binary constraints that limit the number of variables that may be assigned particular values. Counter constraints can be represented by binary constraints, at a cost. We analyse the cost, show how a counter can be represented as a linear number of binary constraints, and demonstrate empirically that even with the optimal reduction,an explicit representation of counters is preferable to their representation as a set of binary constraints. For counter constraints, value ordering is essential. An heuristic for value ordering on constraint satisfaction problems (CSP), based on the estimated likelihoodof a solution, is presented. The proposed value ordering heuristic is useful for counter constraints, as well as for binary CSPs, where it can be used to approximate the number of solutions consistent with a particular value assignment to a variable. The proposed value ordering heuristic integrates counter constraints with binary constraint networks in a novel manner. Counter constraints are problematic for most heuristics, which are local in scope, yet we demonstrated empirically that the proposed value ordering heuristic is significantly superior to heuristics used in previous work.     

一种并行工程约束分解方法

在并行工程产品开发过程中,往往按照问题的结构特点将较大规模的问题分解成一些子问题,并希望通过求解子问题来获得原问题的解。实际中,分解得到的子问题之间往往不是完全独立的,一般的简单分解方法只能有限地降低求解难度和简化问题规模。如何进一步分解各个子问题间的关系,使各个子问题的设计结果不但满足原问题的总体要求而且还能由此获得优化的总体设计结果是一个重要问题。该文给出了分解的意义,提出了基于约束的优化分解方法。     

Solving Mixed and Conditional Constraint Satisfaction Problems

Constraints are a powerful general paradigm for representing knowledge in intelligent systems. The standard constraint satisfaction paradigm involves variables over a discrete value domain and constraints which restrict the solutions to allowed value combinations. This standard paradigm is inapplicable to problems which are either:(a) mixed, involving both numeric and discrete variables, or(b) conditional,¹ containing variables whose existence depends on the values chosen for other variables, or(c) both, conditional and mixed.We present a general formalism which handles both exceptions in an integral search framework. We solve conditional problems by analyzing dependencies between constraints that enable us to directly compute all possible configurations of the CSP rather than discovering them during search. For mixed problems, we present an enumeration scheme that integrates numeric variables with discrete ones in a single search process. Both techniques take advantage of enhanced propagation rule for numeric variables that results in tighter labelings than the algorithms commonly used. From real world examples in configuration and design, we identify several types of mixed constraints, i.e. constraints defined over numeric and discrete variables, and propose new propagation rules in order to take advantage of these constraints during problem solving.