Tee Yong Chew, Martin Henz, and Ka Boon Ng. A toolkit for constraint-based inference engines. In Practical Aspects of Declarative Languages, Second International Workshop, LNCS, Volume 1753, pages 185--199, Boston, MA, January 2000. Springer-Verlag.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....a custom made search engine that has the features required by a particular application. While increasing usability, this idea can help to understand what are the minimal abstractions needed to support a particular feature of a search engine. 155 Chew, Henz, and Ng describe a search toolkit in [22] that allows the orthogonal combination of some of the features introduced in this thesis. However, the more challenging features such as parallelism and different recomputation strategies are not covered. Reusing Parallel Engines One of the main goals for parallel search engines in Chapter 9 has ....

Tee Yong Chew, Martin Henz, and Ka Boon Ng. A toolkit for constraint-based inference engines. In Pontelli and Costa [108], pages 185--199.

....returns to a previous decision and chooses an alternative. In this section, we present a Java abstract search engine, called JASE , which has been actually designed for JCHR but can be used for any Java constraint library. The design of JASE has been inspired by a proposal for the Figaro system [30]. For the search, an exploration is used to walk through a search tree which consists of nodes ; the behavior of the nodes is implemented by choices. A search tree is an unbalanced binary tree. Each inner node of the tree represents one state of the constraint system. Each edge between two ....

T. Chew, M. Henz, and K. Ng. A toolkit for constraint-based inference engines. In Practical Aspects of Declarative Languages, 2000.

....the search returns to a previous decision and chooses an alternative. Most existing libraries and languages have either only depth rst search, e.g. CHIP [6] or support the programming of di erent search algorithms through special purpose language constructs, e.g. Oz [14] or CLAIRE [2] Figaro [3] was proposed to support programmable search algorithms in a C library by representing constraint stores as data objects. Our work is inspired by the Figaro system. We provide an abstract search engine, called JASE (to pronounce chase , because it chases a solution ) which has been actually ....

....type Object. They are associated with a type and a name. Object X = new Object( cs.addVariable(X, java.lang.Integer , X ) The last step in setting up the constraint system is inserting initial constraints with the addGoalConstraint method of ConstraintSystem. Here, the constraint fdEnu(X,[2,3,4,5,6]) is created and inserted into the constraint store: cs.addGoalConstraint(new FDENUConstraint(X,createList(2,6) Now, the search engine is being set up. In this particular example, the values of the variable X should simply be enumerated. So, a container with the variable is created: ....

T. Chew, M. Henz, and K. Ng. A toolkit for constraint-based inference engines. In Practical Aspects of Declarative Languages, 2000.

....the search returns to a previous decision and chooses an alternative. Most existing libraries and languages have either only depth rst search, e.g. CHIP [4] or support the programming of di erent search algorithms through special purpose language constructs, e.g. Oz [12] or CLAIRE [1] Figaro [2] was proposed to support programmable search algorithms in a C library by representing constraint stores as data objects. Our work is inspired by the Figaro system. We provide an abstract search engine, called JASE (to pronounce chase , Fig. 4. Expansion because it chases a solution ) which ....

....type Object. They are associated with a type and a name. Object X = new Object( cs.addVariable(X, java.lang.Integer , X ) The last step in setting up the constraint system is inserting initial constraints with the addGoalConstraint method of ConstraintSystem. Here, the constraint fdEnu(X,[2,3,4,5,6]) is created and inserted into the constraint store: cs.addGoalConstraint(new FDENUConstraint(X,createList(2,6) Now, the search engine is being set up. In this particular example, the values of the variable X should simply be enumerated. So, a container with the variable is created: ....

T. Chew, M. Henz, and K. Ng. A toolkit for constraint-based inference engines. In Practical Aspects of Declarative Languages, 2000.

No context found.

Tee Yong Chew, Martin Henz, and Ka Boon Ng. A toolkit for constraintbased inference engines. In Enrico Pontelli and Vtor Santos Costa, editors, Practical Aspects of Declarative Languages, Second International Workshop, PADL 2000, Lecture Notes in Computer Science 1753, pages 185-199, Boston, MA, 2000. Springer-Verlag, Berlin.

....called spaces [Sch97,Sch00] SALSA [LC98] is a language designed for specifying search algorithms using the concept of choice points. The Figaro library provides a modular architectures for customizing the state restoration policies, in addition to the branching and exploration of tree search [HMN99,CHN00,Ng01,CHN01]. Among the existing frameworks for tree search, we observe that abstractions are limited to customized exploration, branching and state restoration policies. We argue, however, that as the complexity of application speci c search algorithms increases, it becomes more dicult to describe the ....

....We propose a compositional framework for describing complex search scenarios using engines. Each engine performs a single function (e.g. search, print) and engines can be composed together to form more complex search scenarios. We are designing and implementing the framework using Figaro [HMN99,CHN00,Ng01,CHN01], a C CP library, further pursuing its objective of high reusability and rapid prototyping for search components. Notice that we design our framework within the objectoriented paradigm rather than the logic paradigm. We present the compositional framework in Section 2. Sections 3 and 4 discuss ....

[Article contains additional citation context not shown here]

Tee Yong Chew, Martin Henz, and Ka Boon Ng. A toolkit for constraintbased inference engines. In Enrico Pontelli and Vtor Santos Costa, editors, Practical Aspects of Declarative Languages, Second International Workshop, PADL 2000, Lecture Notes in Computer Science 1753, pages 185-199, Boston, MA, 2000. Springer-Verlag, Berlin.

....and Perspectives The design presented here has been inspired by a proposal for an ML library [HS99] in which data structures for representing constraint stores were called rooms. The design presented here has been used to develop a modular architecture for programming search algorithms [CHN00] In this architecture, aspects of search algorithms such as the memory policy, optimization, interactivity and search tree visualization can be programmed independently. The goal to allow the programming of di erent search algorithms in a library for nite domain constraint programming has been ....

Tee Yong Chew, Martin Henz, and Ka Boon Ng. A toolkit for constraintbased inference engines. In Practical Aspects of Declarative Languages, Second International Workshop, PADL'00, Lecture Notes in Computer Science. Springer-Verlag, 2000. to appear.

....and Perspectives The design presented here has been inspired by a proposal for an ML library [HS99] in which data structures for representing constraint stores were called rooms. The design presented here has been used to develop a modular architecture for programming search algorithms [CHN00] In this architecture, aspects of search algorithms such as the memory policy, optimization, interactivity and search tree visualization can be programmed independently. The goal to allow the programming of different search algorithms in a library for finite domain constraint programming has ....

Tee Yong Chew, Martin Henz, and Ka Boon Ng. A toolkit for constraintbased inference engines. In Practical Aspects of Declarative Languages, Second International Workshop, PADL'00, Lecture Notes in Computer Science. Springer-Verlag, 2000. to appear.

No context found.

Tee Yong Chew, Martin Henz, and Ka Boon Ng. A toolkit for constraint-based inference engines. In Practical Aspects of Declarative Languages, Second International Workshop, LNCS, Volume 1753, pages 185--199, Boston, MA, January 2000. Springer-Verlag.

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