A. Aggoun and N. Beldiceanu. Overview of the CHIP Compiler System. In F. Benhamou and A. Colmerauer, editors, Constraint Logic Programming. Selected Research, pages 421--435. MIT Press, 1993.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....enhanced with various new constraints such as constructive disjunction, boolean constraints, non linear constraints and symbolic constraints. 1. INTRODUCTION Constraint Logic Programming (CLP) has shown to be a very active eld of research over recent years, and languages such as CHIP [21, 44, 2], CLP(R) 26, 28] and PrologIII [17] have proved that this approach opens Logic Programming (LP) up to a wide range of real life applications. Address correspondence toPhilippe Codognet and Daniel Diaz, INRIA Rocquencourt, BP 105, 78153 Le Chesnay, FRANCE. email: ....

....77 disjunctions of 2 linear inequations. On the linear equation problems the lowest speedup factor is 2 with peaks reaching 8. On the other programs clp(FD) is around 4 times faster than CHIP. We can also compare clp(FD) with the CHIP compiler system. This compiler is not publicly available but [2] presents execution times for the queens and the bridge programs on a Sparc 1 (18 Mips) We have thus normalized this times by a factor 1=1:6 to obtain a fair comparison with our timings measured on a Sparc 28 Gain Analysis crypta 8919 5047 43.41 22.67 14.80 26.93 alpha 904936 518793 ....

A. Aggoun and N. Beldiceanu. Overview of the CHIP Compiler System. In 8th International Conference of Logic Programming, Paris, France, MIT Press, 1991. Also in Constraint Logic Programming: Selected Research, A. Colmerauer and F. Benhamou (Eds.). MIT Press, 1993.

....is given, along with an example showing a use of the method. 1 Introduction Since its introduction a decade ago, the Constraint Logic Programming paradigm has raised up a growing interest among researchers as well as industrial circles. Many implementations are by now available (e.g. CHIP [1], Prolog IV [5] ECLiPSe, clp(BNR) 4] clp(FD) 6] and have been used to solve satisfactorily hard problems. By taking the best of two worlds (declarativity and expressiveness of Logic Programming, and efficiency of the algorithms of constraint Solving and Operational Research) CLP has ....

A. Aggoun and N. Beldiceanu. Overview of the CHIP Compiler System. In F. Benhamou and A. Colmerauer, editors, Constraint Logic Programming: Selected Research, pages 421--435. MIT Press, 1993.

....for reconstruction are trailing and recomputation. While recomputation computes everything from scratch, trailing records for each state changing operation the information necessary to undo its e#ect. Most current constraint programming systems are trailing based. Many of them, for example CHIP [2], cc(FD) 13] Eclipse [3] and clp(FD) 5] are built on top of Prolog, which itself is trailing based. But also systems that are not built on top of Prolog, like Screamer [12] Lisp) and ILOG Solver [6] C ) use trailing. Appears in: Danny De Schreye, editor, Proceedings of the Sixteenth ....

....that if the contents of n 1 successive locations are changed, n 2 entries are added to the trail: one for the first location s address, a second entry for n, and n entries for the locations values. For a discussion of time stamps and a multiple value trail in the context of the CHIP system, see [1, 2]. A general but brief discussion of issues related to implementation issues for trailing based constraint programming systems can be found in [7] Trailing requires that all operations are search aware: search is not an orthogonal issue to the rest of the system. Complexity in design and ....

Abderrahamane Aggoun and Nicolas Beldiceanu. Overview of the CHIP Compiler System. In Frederic Benhamou and Alain Colmerauer, editors, Constraint Logic Programming: Selected Research, pages 421--437. The MIT Press, Cambridge, MA, USA, 1993.

....constraints fall into this category and usually they form the bulk of constraints used in modelling a problem. More complicated propagation constraints are nonlinear constraints, usually restricted to a few forms like X Y = Z and X 2 = Z. Modern propagation based constraint solvers like CHIP [1, 11] and ILOG SOLVER [10] also include libraries of specialized propagation methods for complex constraints such as cumulative and cycle. Because of the uniform communication mechanism complex multi faceted combinatorial problems can be handled where di erent subparts are solved eciently by ....

A. Aggoun and N. Beldiceanu. Overview of the CHIP compiler system. In F. Benhamou and A. Colmerauer, editors, Constraint Logic Programming: Selected Research, pages 421-435. MIT Press, 1993.

....[23] have demonstrated the validity of the CLP approach and its ability to solve even large real life problems. As far as the implementation of CLP systems is concerned, a variety of different approaches has been already tried: using a Prolog system with coroutining [26] using a dedicated system [2] and or defining extensions to the WAM [16, 7] Following the CLP semantics described in [15] the dedicated CLP systems usually make a difference between constraints on one side and normal Prolog predicates on the other side. The constraints and their propagation are hardcoded in the ....

....of the conference we should already have a working system ready and so it should be possible to include also the parallel figures. 9 5 CLP(FD) Example 5. 1 Implementation The implementation of variables with finite integer domains is also quite straightforward, we follow well known principles [2, 7]. The FD extension uses the following metaterm attribute declaration: metaattribute(domain, unify: unifydomain 2, testunify: testunifydomain 2, compareinstances: compareinstancesdomain 3, copyterm: copytermdomain 2 ] The unification handler is similar to the one described in [20] when a ....

[Article contains additional citation context not shown here]

Abderrahmane Aggoun and Nicolas Beldiceanu. Overview of the CHIP compiler system. In Koichi Furukawa, editor, Proceedings of the Eighth International Conference on Logic Programming, pages 775--789, Paris, France, 1991. The MIT Press.

....other agents. fl To support prototyping, development and use of various system extensions that use different control schemes and to ensure that various extensions can work together in one system. Previous ECRC systems have used special data types, like suspending variables [19] domain variables [1] or consecutively updatable fields [26] to implement 1 advanced control structures, used primarily for CLP implementations. These data structures and their processing were implemented at low level, with no or only limited Prolog access. Our experience from these systems has led us to a scheme ....

....a finite domain variable which occurs in a linear rational constraint. These requirements are satisfied by the multiple attribute scheme based on [3] which allows to define variable attributes independently. This scheme has an obvious advantage over systems which are based on a single attribute [13, 1, 26]: They have to use a different structure (or data with different tag) for every extension in the system. Sharing of different attributes in one variable is achieved by defining new structures or tags for every possible combination. For each such new combination it is also necessary to write code ....

[Article contains additional citation context not shown here]

Abderrahmane Aggoun and Nicolas Beldiceanu. Overview of the CHIP compiler system. In Koichi Furukawa, editor, Proceedings of the Eighth International Conference on Logic Programming, pages 775--789, Paris, France, 1991. The MIT Press.

....and solutions. The aim of this paper is to extend the local consistency algorithms, de ned in [4, 6] for SCSPs, to capture some approximated algorithms which have been proved useful for classical constraints and current constraint programming systems like clp(FD) 7] Ilog Solver [14] and CHIP[1]. The SCSP framework comes with a class of local consistency (LC) algorithms, that are useful to perform a simpli cation of the problem (and that can also be used during its labeling) cutting away some tuples of domain values that are shown to be useless w.r.t. its solutions. This class of ....

A. Aggoun and N. Beldiceanu. Overview of the CHIP Compiler System. MIT Press, 1993.

....Some elements of the runtime structure will also be mentioned. In general, the essential features of the parts of an abstract machine dealing with constraints will differ greatly over CLP languages using different constraint domains. This is exemplified in the literature on CLP(R) 135] CHIP [3], and CLP(FD) 75] The following presentation, though based on one work [135] contains material that is relevant to abstract machines for many CLP languages. We begin by arguing that an abstract machine is the right approach in the first place. Abstract machines have been used for implementing ....

A. Aggoun & N. Beldiceanu, Overview of the CHIP Compiler System, in: Constraint Logic Programming: Selected Research, F. Benhamou and A. Colmerauer (Eds.), MIT Press, 421--435, 1993.

....kind of search algorithm is given in Section 4.6. It is interesting to note that the trail only starts trailing after the first mark is called. This is because this is the earliest state of the room that the programmer intends to backtrack to. In fact, this trail uses a time stamping technique (Aggoun Beldiceanu, 1991), which prevents unnecessary trailing. What happens is that in each of the variables and propagators, there is a stamp to indicate when the object is last trailed. In the room that maintains the trail, there is also a stamp to indicate the current mark frame. If a variable or propagator is already ....

Aggoun, A., & Beldiceanu, N. (1991). Overview of the CHIP Compiler System. In Eighth International ConferenceonLogic Programming (ICLP '91) (p. 775-789). Paris, France: MIT Press.

....the future, a major requirement from a CLP system will be the ability to easily incorporate a new constraint solving algorithm. This trend is underlined by the increasing number of algorithms (besides standard unification) which have been used in CLP systems: linear equality and inequality solvers [1, 17], non linear constraint solvers [14] linear Diophantine constraint solvers [7] finite domain constraint solvers [1, 8] finite set constraint solvers [11] interval solvers [2, 6] neural networks [21] etc. A first problem is that the implementations of future CLP systems must be open to new ....

.... This trend is underlined by the increasing number of algorithms (besides standard unification) which have been used in CLP systems: linear equality and inequality solvers [1, 17] non linear constraint solvers [14] linear Diophantine constraint solvers [7] finite domain constraint solvers [1, 8], finite set constraint solvers [11] interval solvers [2, 6] neural networks [21] etc. A first problem is that the implementations of future CLP systems must be open to new algorithms. A second problem is how to efficiently implement a constraint solving algorithm in a CLP system. By ....

A. Aggoun and N. Beldiceanu. Overview of the CHIP compiler system. In K. Furukawa, editor, Proceedings of the Eighth International Conference on Logic Programming, pages 775--789, Paris, France, 1991. The MIT Press. 9

....with the three major approaches (other than our own) for disjunctive and hypothetical reasoning over constraints. CHAPTER 1. INTRODUCTION 33 1.2. 1 The cc(FD) Language The cc(FD) 20 programming language was recently introduced in [26, 23, 21] as a rational reconstruction of the CHiP language [17, 1]. Although described as a concurrent constraint language, it is closer to the CLP(D) approach in that the store is not maintained strongly K consistent, but rather uses an arc consistency algorithm [25] Thus the store may at times be inconsistent. If inconsistency is discovered, backtracking is ....

....c) where each A i is a propositional letter. This can be read as the logical implication V n i=1 A i c. This work builds on the assumption based truth maintenance system (ATMS) of de Kleer [14] and the general notions of boolean constraint systems introduced in Prolog III [12, 13] and CHiP [17, 1]. It may be viewed as a generalization of the ATMS and a formalization of the some uses of boolean variables to combine constraints in logic programming. Alternately, it may be seen as a restricted form of conditional constraints. The formal semantics of this system are straightforward. They are ....

[Article contains additional citation context not shown here]

Abderrahmane Aggoun and Nicolas Beldiceanu. Overview of the CHIP compiler system. In Furukawa [18], pages 775--789.

....more interpreted domains of constraints. CLP is parameterised by the domain of discourse D over which the constraints range. Satisfiability in D of a set of constraints is tested by algorithms specialised to the domain. Many CLP languages have been implemented, including CLP(R) CHIP, and Eclipse [1, 22, 43]. There are a number of survey papers and books available [34, 42, 74, 75] Jaffar and Lassez show that for every domain D with the two properties of solution compactness and satisfaction completeness, CLP(D) has many desirable characteristics. These include the existence of equivalent declarative ....

....ways of modelling a problem using constraints. For example, in finite domain CLP, the day of the week on which an event occurs will probably be represented by a single variable ranging over seven possible values. In integer programming, however, there would be seven variables each with domain [0,1], representing the truth or falsity of a statement such as The event is on Monday . Schemas written on the assumption that the first of these two representations was going to be used would not work for the second. Furthermore, in the case of composition of two self consistent systems which ....

Abderrahmane Aggoun and Nicolas Beldiceanu. Overview of the CHIP Compiler System. In Koichi Furukawa, editor, ICLP'91: Proceedings 8th International Conference on Logic Programming, pages 775--789. MIT Press, 1991. (A longer version is in [2]).

....contain the saved information before the change. Furthermore, we are keeping track of the last mark after which variable and propagator entries are changed so that we do not need to create a change entry if the entry has been changed already; this technique is called time stamping and reported in [AB91] We are using the propagator entries in the trail to implement exhaustive propagation by linking the propagators in form of a queue, using their next entry. More specifically, the type declaration in Program 5 describes the trail entries. In this implementation, the types mark, var and prop are ....

Abderrahamane Aggoun and Nicolas Beldiceanu. Overview of the CHIP Compiler System. In Koichi Furukawa, editor, Logic Programming: Proceedings of the Eight International Conference, pages 775--788, Paris, France, June 1991. The MIT Press.

....is given, along with an example showing a use of the method. 1 Introduction Since its introduction a decade ago, the Constraint Logic Programming paradigm has raised up a growing interest among researchers as well as industrial circles. Many implementations are by now available (e.g. CHIP [1], Prolog IV [5] ECLiPSe, clp(BNR) 4] clp(FD) 6] and have been used to solve satisfactorily hard problems. By taking the best of two worlds (declarativity and expressiveness of Logic Programming, and efficiency of the algorithms of constraint Solving and Operational Research) CLP has ....

A. Aggoun and N. Beldiceanu. Overview of the CHIP Compiler System. In F. Benhamou and A. Colmerauer, editors, Constraint Logic Programming: Selected Research, pages 421--435. MIT Press, 1993.

....being very simple and efficient to handle at the implementation level. For instance, the clp(FD) system, a constraint logic programming language based on FD constraints, shows that flexibility and efficiency can indeed be combined, as it is on average several times faster than the CHIP system [1]. A domain in FD is a (non empty) finite set of natural numbers (i.e. a range) More precisely a range is a subset of f0; 1; infinityg where infinity is a particular integer denoting the greatest value that a variable can take. We use the interval notation k 1 : k 2 as a shorthand for the ....

A. Aggoun and N. Beldiceanu. Overview of the CHIP Compiler System. In Constraint Logic Programming: Selected Research, A. Colmerauer and F. Benhamou (Eds.). MIT Press, 1993.

....say that the interval (3; 6] is associated with the variable X. Semantically, an interval variable is an ordinary logical variable. We distinguish interval variables from logical variables purely for implementation efficiency. Resembling domain variables in finite domain languages [52] e.g. CHIP [1, 19], an interval variable is represented as a variable with an associated interval. Its heap representation is shown in figure 5, where ITAG is a new tag introduced for interval variables. Each interval variable in CIAL keeps a list of constraints in which the variable appears. When an interval ....

A. Aggoun and N. Beldiceanu. Overview of the CHIP compiler system. In Proceedings of the Eighth International Conference on Logic Programming, pages 775--789, Paris, France, 1991.

....say that the interval (3; 6] is associated with the variable X. Semantically, an interval variable is an ordinary logical variable. We distinguish interval variables from logical variables purely for implementation efficiency. Resembling domain variables in finite domain languages [57] e.g. CHIP [1, 19], an interval variable is represented as a variable with an associated interval. Its heap representation is shown in figure 4.2, where ITAG is a new tag introduced for interval variables. Each interval variable in CIAL keeps a list of constraints in which the variable appears. When an interval ....

A. Aggoun and N. Beldiceanu. Overview of the CHIP compiler system. In Proceedings of the Eighth International Conference on Logic Programming, Paris, France, 1991.

No context found.

A. Aggoun and N. Beldiceanu. Overview of the CHIP Compiler System. In F. Benhamou and A. Colmerauer, editors, Constraint Logic Programming. Selected Research, pages 421--435. MIT Press, 1993.

No context found.

A. Aggoun and N. Beldiceanu. Overview of the CHIP Compiler System. MIT Press, 1993.

No context found.

Aggoun, A., and Beldiceanu, N. Overview of the CHIP compiler system. In ICLP'91: Proceedings 8th International Conference on Logic Programming (Paris, 1991), K. Furukawa, Ed., MIT Press, pp. 775--789.

No context found.

A. Aggoun and N. Beldiceanu. Overview of the CHIP Compiler System. In 8th International Conference of Logic Programming, Paris, France, MIT Press 1991.

No context found.

A. Aggoun and N. Beldiceanu. Overview of the CHIP compiler system. In F. Benhamou and A. Colmerauer, editors, Constraint Logic Programming: Selected Research, pages 421-435. MIT Press, 1993.

No context found.

A. Aggoun & N. Beldiceanu, Overview of the CHIP Compiler System, in: Constraint Logic Programming: Selected Research, F. Benhamou and A. Colmerauer (Eds.), MIT Press, 421--435, 1993.

No context found.

A. Aggoun and N. Beldiceanu. Overview of the CHIP compiler system. In Koichi Furukawa, editor, Proc. 8th International Conference on Logic Programming, pages 775--789, Paris, France, June 1991. MIT Press.

No context found.

Abderrahmane Aggoun and Nicolas Beldiceanu. Overview of the CHIP compiler system. In Koichi Furukawa, editor, ICLP'91: Proceedings 8th International Conference on Logic Programming, pages 775--789, Paris, June 1991. MIT Press.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC