Sidebottom, G. (1993). A Language for Optimizing Constraint Propagation. PhD Thesis, Simon Fraser University.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....in CHIP [D 88] allow the definition of propagation of constraints in a limited way. ffl Constraint combinators in cc(FD) VH91] allow to build more complex constraints from simpler constraints. ffl Constraints connected to a Boolean variable in BNR Prolog [BeOl92] and nested constraints [Sid93] allow to express any logical formula over primitive constraints. ffl Indexicals in clp(FD) CoDi93] allow to implement constraints over finite domains at a medium level of abstraction. The work was performed while visiting CWG at LMU with financial support from DFG. y Constraint Working Group ....

G.A. Sidebottom, A Language for Optimizing Constraint Propagation, 1993, Simon Fraser University, Canada.

....allow defining propagation of constraints in a limited way (Section 3) ffl Constraint combinators, cc(FD) vH91] allow building more complex constraints from simpler constraints (see also Section 8. 1) ffl Constraints connected to a Boolean variable, BNR Prolog [BeOl92] nested constraints [Sid93], allow expressing any logical formula over primitive constraints. ffl Indexicals, clp(FD) CoDi96] allow implementing constraints over finite domains at a medium level of abstraction. ffl Meta and attributed variables [Hol92] allow attaching constraints to variables (Section 7) It should ....

G. A. Sidebottom, A Language for Optimizing Constraint Propagation, Thesis, Simon Fraser University, Canada, 1993.

....arithmetic in the constraint reasoning sys 1 Prolog III provides the option of using floating point arithmetic, although the default is rational arithmetic. tem Echidna [24] Based on hierarchical consistency techniques [39] Echidna can handle unions of disjoint intervals. Sidebottom [49] describes the use of projection constraints for compiling and optimizing constraint propagation in the numeric and Boolean domains. He shows that all the constraints available in CLP(BNR) can be directly expressed by projection constraints without applying constraint decomposition. Also, the user ....

G.A. Sidebottom. A Language for Optimizing Constraint Propagation. PhD thesis, School of Computer Science, Simon Fraser University, November 1993.

....the feasible values for X 1 ; X i Gamma1 ; X i 1 ; Xn . The basic idea is to express C as n indexicals, one for each X i , encoding a local consistency method for solving C. Each indexical is a projection of C onto X i ; hence, indexicals are also known as projection constraints [21], and have been used in several implementations [9,21,6] Ranges are defined using a constraint programming assembly language , which gives the programmer precise control over the level of consistency, and can yield more efficient solutions than relying on the solver s built in constraints. An ....

....X i 1 ; Xn . The basic idea is to express C as n indexicals, one for each X i , encoding a local consistency method for solving C. Each indexical is a projection of C onto X i ; hence, indexicals are also known as projection constraints [21] and have been used in several implementations [9,21,6]. Ranges are defined using a constraint programming assembly language , which gives the programmer precise control over the level of consistency, and can yield more efficient solutions than relying on the solver s built in constraints. An important feature of any constraint solver is ....

[Article contains additional citation context not shown here]

Gregory Sidebottom. A Language for Optimizing Constraint Propagation. PhD thesis, Simon Fraser University, November 1993.

....example, FD.bool Ls constrains all elements of the list Ls to be between 0 and 1. If we bind Z to 1, W is also bound to 1. The constraints X=Y Y=1 lead to the same result. As an example for the use of boolean constraints we consider the problem of hardware diagnosis (the example is taken from [10]) To an existing piece of hardware we assume the input and output values as given. We now want to know, which gates in the hardware have to be defective to produce the given behavior. We assume a fixed number of defective gates. The hardware we consider is an n bit adder consisting of n ....

....RO fi FD.fd SY Sum Coord Xr Sr SS SY =# B S SS end [ nil#nil then SY=0 end end Now we can add the additional applications Capacity XCoord Sizes SX SY Capacity YCoord Sizes SY SX into the procedure Square to obtain a better run time. As another example consider the values Ss=[18 15 14 10 9 8 7 4 1] SX=32 SY=33 where the introduction of redundant constraints is crucial. In Section 2.2 we have seen how the incompleteness due to local propagation may be alleviated by using search. Consider the following program. declare proc Diff Ls N List N 1 Ls FD.fromTo Ls 1 N FD.allDifferent Ls ....

G.A. Sidebottom. A Language for Optimizing Constraint Propagation. PhD thesis, Simon Fraser University, Canada, 1993.

No context found.

Sidebottom, G. (1993). A Language for Optimizing Constraint Propagation. PhD Thesis, Simon Fraser University.

No context found.

G. A. Sidebottom, A Language for Optimizing Constraint Propagation, Thesis, Simon Fraser University, Canada, 1993.

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