P. V. Hentenryck and Y. Deville. Operational semantics of constraint logic programming over finite domains. In J. Ma/luszy'nski and M. Wirsing, editors, PLILP91, number 528 in LNCS, pages 395--406. Springer-Verlag, aug 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....of concrete starting times. It turns out that a valid schedule can be obtained just by simply assigning the starting times to the minimal value of those still remaining. This means that no further search is needed. This observation can be gathered from a similar theorem of Van Hentenryck [8]. This theorem essentially states that mere constraint propagation is indeed sufficient for detecting any inconsistency if there are only constraints of the form x c y and x c = y such that c is an arbitrary integer. The theorem moreover states that a solution can be obtained in a way very ....

HENTENRYCK, P. V., AND DEVILLE, Y. Operational semantics of constraint logic programming over finite domains. In Proceedings of the AAAI Spring Symposium Series (1991), pp. 128--146.

....; p 11 ) where c 1 = fa,c,g,ug, c 11 = fa,c,g,ug, p 1 0, p 2 = p 1 1, p 11 = p 10 1, char comp(c 1 ,c 11 ) char comp(c 4 ,c 8 ) c 6 = c, where char comp defines character complement over the RNA alphabet. We have chosen constraint logic programming over finite domains [30] as a paradigm for implementation because of the declarative nature of our structure language and the use which it makes of finite domain constraints. In our implementation sequences are represented as lists, and thus string variables comprise lists whose elements are pairs of (Chars,Pos) We ....

P. V. Hentenryck and Y. Deville. Operational semantics of constraint logic programming over finite domains. In J. Ma/luszy'nski and M. Wirsing, editors, PLILP91, number 528 in LNCS, pages 395--406. Springer-Verlag, aug 1991.

....Z. Predicate indomain assigns values to variables from their domain. In this case it is not needed since the active use of the constraints resulted to a unique value for each variable. The techniques used in CHIP are forward checking, partial and full lookahead and are described in detail in [Hen89, MD88, HD90]. In the above case forward checking was applied leaving us with a backtrack free search. The same problem solved with a generate and test paradigm in PROLOG, like the one of example 1.1.1, would lead to a possible search space of 60 nodes (worst case) CHAPTER 1. INTRODUCTION 15 The efficiency ....

P.V. Hentenryck and Y. Deville. Operational Semantics of Constraint Logic Programming over Finite Domains. Technical report no. cs-90-23, Brown University, 1990.

....the domain of Y (X) or by the minimum and maximum of Y (X) The domain approximation performs stronger propagation than the interval approximation, but the interval version is more efficient to compute. For a careful examination of domain and interval approximations of constraints see elsewhere [6, 13, 14]. Note that operationally the constraint propagation implemented by the FD constraints may be weaker than what can be performed by a constraint solver for the n ary constraint. Example 2. The constraint X 6= Y can be defined either as X 6= Y : Gamma X in Gamma dom(Y ) Y in Gamma dom(X) ....

P. van Hentenryck and Y. Deville. Operational Semantics of Constraint Logic Programming over Finite Domains. In Proceedings of the 3rd Int. Symposium on Programming Language Implementation and Logic Programming, 1991. 15

....4 1 b 2 3 CLP j 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 2.4 2.5 2.6 2.7 2. 8 k k i j i i n n n n W n n n n n n goal constraint solver state [ f g ffl ffl f g f g f j 2 2 g f 2 j 6 g f j 2 g f g 9 X X X X C C h i Definition Definition Definition Definition Definition In [28] a structural operational approach is also used to define the semantics for a class of CLP languages. However our approach is different as our algorithms are incremental. V V V V x =t ; x =t t i ; k x j ; k x t i ; k x =t ; x =t x t ; x t ffl W V W ....

....the constraint is not but also when it has a unique mgu over that has already been found and it tries to obtain another solution. However, for some classes of problems, an incomplete but efficient constraint solver might turn out to be valuable from a programming standpoint, as pointed out in [28]. The following definitions instantiate definitions 2.9, 2.10 and 2.11 to the case of CLP( programs. 14 b b b c 0 0 0 0 j [ j j [ n 3.11 3.12 2 H E H E H E 0 0 0 3 3 3 2 2 2 3 3 3 2 2 2 CLP = i CLP = CLP = n 0 1 Knapsack problem ( 0 0 ( 4 1 2 0 1 0 2 1 0 5 c var G E E var ....

P. Van Hentenryck and Y. Deville. Operational Semantics of Constraint Logic Programming over Finite Domains. In J. Maluszy'nski and M. Wirsing, editors, , volume 528 of , pages 395--406. Springer-Verlag, Berlin, 1991.

....The former introduces a committed choice aspect into the language, whereas the latter is a variant of the second approach. All these approaches originated for conventional logic programs, but the ideas lift to constraint logic programs, and there are now several proposals based on these ideas [137, 234, 11, 113, 116]. One potential problem with using guarded rules is that the completeness of the operational semantics with respect to the logical semantics of the program can be lost. This incompleteness was shown to be avoided in ALPS [173] modulo infinitely delayed atoms) but that work was heavily reliant ....

P. van Hentenryck & Y. Deville, Operational Semantics of Constraint Logic Programming over Finite Domains, Proc. Symp. on Programming Language Implementation and Logic Programming, LNCS 528, 395--406, 1991.

....As in CLP (X) interpreted constraints are treated by telling them immediately to the constraint store. The special feature of GP (X) is the handling of propagation constraints using generalised propagation. Generalised propagation can be seen as an example of the relaxed tell operation of [HD91] which is discussed in more detail in section 6.3, below. Approximation constraints are extracted from the propagation constraint and they are told to the store. However 2 In [SKL90] each store is viewed as a set. However this second view can only treat distinct tokens of information which ....

....follows that A is consistent with S AC. Moreover if (S A) C, then a fortiori (S AC A) C. Consequently every answer to C with S is also an answer to C with prop(S; C) Therefore the result of propagation on C with prop(S; C) remains prop(S; C) This condition is not satisfied by relaxed tell [HD91] which is an abstraction of generalised propagation (see below 6.3) 5 Some Instances of GP(X) Two implementations of generalised propagation over the Herbrand universe have been completed. In the two following sections the examples we describe have all been run on a GP(HU) implementation ....

[Article contains additional citation context not shown here]

P. Van Hentenryck and Y. Deville. Operational semantics of constraint logic programming over finite domains. In Proc. PLILP'91, Passau, Germany, Aug 1991.

....9 ; p 9 ) where c 1 = fa,c,g,ug, c 9 = fa,c,g,ug, p 1 0; p 2 = p 1 1; p 9 = p 8 1, char comp(c 1 ,c 9 ) char comp(c 4 ,c 6 ) c 5 = c where char comp defines character complement over the RNA alphabet. We have chosen constraint logic programming over finite domains [HD91b] as a paradigm for implementation because of the declarative nature of our structure language and the use which it makes of finite domain constraints. In our implementation sequences are represented as lists, and thus string variables comprise lists whose elements are pairs of (Chars,Pos) We ....

P. V. Hentenryck and Y. Deville. Operational semantics of constraint logic programming over finite domains. In J. Ma/luszy'nski and M. Wirsing, editors, PLILP91, number 528 in LNCS, pages 395--406. Springer-Verlag, aug 1991.

....(glb) and least upper bound (lub) The inclusion and disjointness constraints over these domain variables are solved by applying consistency techniques which allow to perform deterministic computations until we reach a fixed point. This approach can be seen as an adaptation of finite domains [3][9] to finite set domains where the number of elements of the domain is no longer linear but exponential in the size of the upper bound and where the order relation is not total ( but partial ( As a consequence, checking the consistency of arithmetic constraints over each value of a domain would ....

....not very surprising for it is very close to the problem of handling disjunctions in Prolog. 4.2.1 Basic constraints P = P 1 P 2 P denotes the system of basic constraints composed of set constraints and arithmetic constraints. P 2 is the set of basic arithmetic constraints defined in [9] (fax = by c; ax 6= c; ax by c; ax by c; x 2 fa 1 ; a n gg where the a,b,c,a 1 , a n are positive integers and x,y are domain variables) We include P 2 in the system as the solver handles finite (integer) domains when dealing with the cardinality operator #. For reasons of space, ....

[Article contains additional citation context not shown here]

P. Van Hentenryck and Y. Deville. Operational Semantics of Constraint Logic Programming over Finite Domains. In Proceedings of PLILP'91, pages 396-- 406, Passau, Germany, Aug. 1991.

....added to CLP languages over finite domains [4] A variable constrained by a finite domain, specifies that this variable can take any value in an integer domain. In order to prune the search space in an a priori way, the efficiency of consistency technics over finite domains has been demonstrated [13]. But finite domains lead to heavy and costly problem definition for some OR problems, especially when there is a large number of variable dependencies (e.g. set partitioning, resource allocation, hamiltonian cycle, scheduling problems, In such cases the use of sets and relations is more ....

P. Van Hentenryck and Yves Deville. Operational Semantics of Constraint Logic Programming over Finite Domains. 3rd International Language Implementation and Logic Programming. Programming Language Implementation and Logic Programming 1991, J. Maluszynski and M. Wirsing Springer Verlag. pp 396-406 Passau, Germany Aug.

....the domain of Y (X) or by the minimum and maximum of Y (X) The domain approximation performs stronger propagation than the interval approximation, but the interval version is more efficient to compute. For a careful examination of domain and interval approximations of constraints see elsewhere [6, 13, 14]. Note that operationally the constraint propagation implemented by the FD constraints may be weaker than what can be performed by a constraint solver for the n ary constraint. Example 2. The constraint X 6= Y can be defined either as X 6= Y : Gamma X in Gamma dom(Y ) Y in Gamma dom(X) ....

P. van Hentenryck and Y. Deville. Operational Semantics of Constraint Logic Programming over Finite Domains. In Proceedings of the 3rd Int. Symposium on Programming Language Implementation and Logic Programming, 1991.

....of the form: GammaB 1 : Bm c 1 : c n , where B 1 ; Bm are atoms and c 1 ; c n are constraints. 2. 3 Operational Semantics of CLP language The operational semantics of a CLP language can be defined either in terms of logical consequences or in an algebraic way [32] (see [10] for a detailed discussion) Definition 2.3 An answer to a CLP goal G is a conjunction of constraints c 1 : c n such that P; T j= 8) c 1 : c n G) or P j= S (8) c 1 : c n G) where P is a constraint logic program, S is a structure, T is the theory axiomatizing ....

P. Van Hentenryck and Deville Y. Operational semantics of constraint logic programming over finite domains. In Proceedings of PLILP'91, 1991.

....of neighbouring members of the sequence in the normally accepted direction of ordering. A (suitably constrained) string variable is thus schema for a structure, and can be instantiated by matching against an input string (see below) We have chosen constraint logic programming over finite domains [14] as a paradigm for implementation because of the declarative nature of our structure language and the use which it makes of finite domain constraints. In our implementation sequences are represented as lists, and thus string variables comprise lists whose elements are pairs of (Chars,Pos) We ....

P. V. Hentenryck and Y. Deville. Operational semantics of constraint logic programming over finite domains. In J. Ma/luszy'nski and M. Wirsing, editors, PLILP91, number 528 in LNCS, pages 395--406. Springer-Verlag, aug 1991.

....1 For a given GP (X) program P , a solution to a query G under a constraint S is an valuation for which X j= S and P j= X G 3.2. 3 Operational Semantics for GP(X) We have chosen a transformational semantics for our constraint logic programming system following the approach of [Sar89] and [HD91] GP(X) States At any point in a GP (X) evaluation, the current state is formalised as a triple fG 1 ; G r g; fA 1 ; A s g; fC 1 ; C t g . The current goal fG 1 ; G r g is a set of atoms, which may include user atoms and atomic constraints. The current set of ....

....if is any solution to A i under store S old , then X j= prop(A i ; S old ) It is the spontaneous production of new information, in the form of approximation constraints, that we call generalised propagation. Generalised propagation can be seen as an example of the relaxed tell operation of [HD91] which is discussed in more detail in section 5.2, below. For any state G; A; S in which A i is a propagation agent (A i 2 A) there is a possible state transition corresponding to single propagation steps on an agent A i in each subset S old of the constraint store S. However if prop(A i ; ....

[Article contains additional citation context not shown here]

P. Van Hentenryck and Y. Deville. Operational semantics of constraint logic programming over finite domains. In Proc. PLILP'91, Passau, Germany, Aug 1991.

.... the set oriented specification method of [SP87] we define a successor constraint on string character positions by 8a : SV : 8i : 1 : len(a Gamma 1) 8x : 1 : mj(x 2 proj2(a i ) x m) succ(x) 2 proj2(a succ(i) We have chosen constraint logic programming over finite domains [HD91b] as a paradigm for implementation because of the declarative nature of our structure language and the use which it makes of finite domain constraints. In our implementation sequences are represented as lists, and thus string variables comprise lists whose elements are pairs of (Chars,Pos) We ....

P. V. Hentenryck and Y. Deville. Operational semantics of constraint logic programming over finite domains. In J. Ma/luszy'nski and M. Wirsing, editors, PLILP91, number 528 in LNCS, pages 395--406. Springer-Verlag, aug 1991.

....of neighbouring members of the sequence in the normally accepted direction of ordering. A (suitably constrained) string variable is thus schema for a structure, and can be instantiated by matching against an input string (see below) We have chosen constraint logic programming over finite domains [19] as a paradigm for implementation because of the declarative nature of our structure language and the use which it makes of finite domain constraints. In our implementation sequences are represented as lists, and thus string variables comprise lists whose elements are pairs of (Chars,Pos) We ....

P. V. Hentenryck and Y. Deville. Operational semantics of constraint logic programming over finite domains. In J. Ma/luszy'nski and M. Wirsing, editors, PLILP91, number 528 in LNCS, pages 395--406. Springer-Verlag, aug 1991.

....from the declarative point of view, we now come to their operational semantics. We develop a general framework, branch and infer, that unifies the classical branch and cut approach from integer linear programming [28] with the usual operational semantics of finite domain constraint programming [36]. 3.1 Primitive and non primitive constraints We start from a common distinction in finite domain constraint programming [36] and split the constraint language L into a set Prim(L) of primitive constraints and a set NPrim(L) of non primitive constraints, such that L = Prim(L) NPrim(L) and ....

....that unifies the classical branch and cut approach from integer linear programming [28] with the usual operational semantics of finite domain constraint programming [36] 3. 1 Primitive and non primitive constraints We start from a common distinction in finite domain constraint programming [36] and split the constraint language L into a set Prim(L) of primitive constraints and a set NPrim(L) of non primitive constraints, such that L = Prim(L) NPrim(L) and Prim(L) NPrim(L) Intuitively, the primitive constraints are those constraints that can be easily solved. In other words, ....

P. van Hentenryck and Y. Deville. Operational semantics of constraint logic programming over finite domains. In Programming language implementation and logic programming, PLILP'91, Passau. Springer, LNCS 528, 1991.

No context found.

P. V. Hentenryck and Y. Deville. Operational semantics of constraint logic programming over finite domains. In J. Ma/luszy'nski and M. Wirsing, editors, PLILP91, number 528 in LNCS, pages 395--406. Springer-Verlag, aug 1991.

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