Results 1  10
of
13
A complexity analysis of spacebounded learning algorithms for the constraint satisfaction problem
 In Proceedings of the Thirteenth National Conference on Artificial Intelligence
, 1996
"... Learning during backtrack search is a spaceintensive process that records information (such as additional constraints) in order to avoid redundant work. In this paper, we analyze the effects of polynomialspacebounded learning on runtime complexity of backtrack search. One spacebounded learning sc ..."
Abstract

Cited by 80 (2 self)
 Add to MetaCart
Learning during backtrack search is a spaceintensive process that records information (such as additional constraints) in order to avoid redundant work. In this paper, we analyze the effects of polynomialspacebounded learning on runtime complexity of backtrack search. One spacebounded learning scheme records only those constraints with limited size, and another records arbitrarily large constraints but deletes those that become irrelevant to the portion of the search space being explored. We find that relevancebounded learning allows better runtime bounds than sizebounded learning on structurally restricted constraint satisfaction problems. Even when restricted to linear space, our relevancebounded learning algorithm has runtime complexity near that of unrestricted (exponential spaceconsuming) learning schemes.
Hybrid backtracking bounded by treedecomposition of constraint networks
 Artificial Intelligence
, 2003
"... We propose a framework for solving CSPs based both on backtracking techniques and on the notion of treedecomposition of the constraint networks. This mixed approach permits us to define a new framework for the enumeration, which we expect that it will benefit from the advantages of two approaches: ..."
Abstract

Cited by 51 (16 self)
 Add to MetaCart
We propose a framework for solving CSPs based both on backtracking techniques and on the notion of treedecomposition of the constraint networks. This mixed approach permits us to define a new framework for the enumeration, which we expect that it will benefit from the advantages of two approaches: a practical efficiency of enumerative algorithms and a warranty of a limited time complexity by an approximation of the treewidth of the constraint networks. Finally, experimental results allow us to show the advantages of this approach. 1
Counting models using connected components
 In Proceedings of the AAAI National Conference
, 2000
"... Recent work by Birnbaum & Lozinskii [1999] demonstrated that a clever yet simple extension of the wellknown DavisPutnam procedure for solving instances of propositional satisfiability yields an efficient scheme for counting the number of satisfying assignments (models). We present a new extension, ..."
Abstract

Cited by 47 (0 self)
 Add to MetaCart
Recent work by Birnbaum & Lozinskii [1999] demonstrated that a clever yet simple extension of the wellknown DavisPutnam procedure for solving instances of propositional satisfiability yields an efficient scheme for counting the number of satisfying assignments (models). We present a new extension, based on recursively identifying connected constraintgraph components, that substantially improves counting performance on random 3SAT instances as well as benchmark instances from the SATLIB and Beijing suites. In addition, from a structurebased perspective of worstcase complexity, while polynomial time satisfiability checking is known to require only a backtrack search algorithm enhanced with nogood learning, we show that polynomial time counting using backtrack search requires an additional enhancement: good learning.
On the spacetime tradeoff in solving constraint satisfaction problems
 in: Fourteenth International Joint Conference on Artificial Intelligence (IJCAI
, 1995
"... A common technique for bounding the runtime required to solve a constraint satisfaction problem is to exploit the structure of the problem's constraint graph [Dechter, 92]. We show that a simple structurebased technique with a minimal space requirement, pseudotree search [Freuder & Quinn, 85], is ..."
Abstract

Cited by 32 (2 self)
 Add to MetaCart
A common technique for bounding the runtime required to solve a constraint satisfaction problem is to exploit the structure of the problem's constraint graph [Dechter, 92]. We show that a simple structurebased technique with a minimal space requirement, pseudotree search [Freuder & Quinn, 85], is capable of bounding runtime almost as effectively as the best exponential spaceconsuming schemes. Specifically, if we let n denote the number of variables in the problem, w * denote the exponent in the complexity function of the best structurebased techniques, and h denote the exponent from pseudotree search, we show h < {w * + 1) (lg(n) + 1). The result should allow reductions in the amount of realtime accessible memory required for predicting runtime when solving CSP equivalent problems. 1
Processing Queries for FirstFew Answers
 Proc. 3rd CIKM Conf
, 1996
"... Special support for quickly finding the firstfew answers of a query is already appearing in commercial database systems. This support is useful in active databases, when dealing with potentially unmanageable query results, and as a declarative alternative to navigational techniques. In this paper, ..."
Abstract

Cited by 28 (2 self)
 Add to MetaCart
Special support for quickly finding the firstfew answers of a query is already appearing in commercial database systems. This support is useful in active databases, when dealing with potentially unmanageable query results, and as a declarative alternative to navigational techniques. In this paper, we discuss query processing techniques for firstanswer queries. We provide a method for predicting the cost of a firstanswer query plan under an execution model that attempts to reduce wasted effort in join pipelining. We define new statistics necessary for accurate cost prediction, and discuss techniques for obtaining the statistics through traditional statistical measures (e.g. selectivity) and semantic data properties commonly specified through modern OODB and relational schemas. The proposed techniques also apply to allanswer query processing when optimizing for fast delivery of the initial query results. 1 Introduction Traditional methods for query processing, primarily those based ...
Using Symmetry of Global Constraints to Speed up the Resolution of Constraint Satisfaction Problems
 in Workshop on Non Binary Constraints, ECAI98
, 1998
"... Abstract. Symmetry in constraint satisfaction problems (CSP) can be used to either compute only a subset of the total solution set, or to prune branches of the search tree. However, detecting symmetry in general is a difficult task. In this paper, we address the problem of detecting and exploiting a ..."
Abstract

Cited by 12 (0 self)
 Add to MetaCart
Abstract. Symmetry in constraint satisfaction problems (CSP) can be used to either compute only a subset of the total solution set, or to prune branches of the search tree. However, detecting symmetry in general is a difficult task. In this paper, we address the problem of detecting and exploiting a particular class of symmetry called intensional permutability, which is based on the notion of interchangeability between variables and can be detected with a very small overhead. This kind of symmetry is detected by collecting information on symmetrical properties of individual constraints. This method works particularly well on problems designed using global constraints. We show how intensional permutability dramatically reduces the search tree for some problems. We propose a simple method to exploit it, which can be implemented as a lightweight extension to most resolution algorithms based on backtracking. We illustrate the method on several symmetrical problems, such as a classical layout problem and the pigeonhole problem, stated with a global constraint. Finally, we extend the method to symmetries involving groups of variables. 1
Loop Optimizations for Acyclic ObjectOriented Queries
 Proceedings of the 1992 Symposium on Interactive 3D Graphics, ACM
"... Nested loop execution of objectoriented queries retains the promise of maintaining the full generality of the object paradigm, independent of the specifics of any single object model. Thus, from this starting point we have developed an objectoriented query optimizer and execution engine. The metho ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
Nested loop execution of objectoriented queries retains the promise of maintaining the full generality of the object paradigm, independent of the specifics of any single object model. Thus, from this starting point we have developed an objectoriented query optimizer and execution engine. The methods, developed to date for only acyclic queries, augment nested loops structures with a simple marking mechanism such that unnecessary loop iterations are not repeated. In the case of acyclic queries, the executions are asymptotically optimal. In contrast to optimal query methods based on semijoin reductions our method involves no preprocessing step and thus avoids the extra I/O associated with semijoins and prevents the formal benefits of semijoin reduction from appearing as a practical improvement. Empirical results comparing our query environment with a commercially available product demonstrate significant performance improvement. 1 Introduction We develop a technique for the optimizatio...
A Prolog Technique Of Implementing Search Of A/o Graphs With Constraints
, 1997
"... . Our research has been motivated by the task of forming a solution subgraph which satisfies given constraints. The problem is represented by an A=O graph. Our approach is to apply a suitably modified technique of dependencydirected backtracking. We present our formulation of the standard chronolog ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
. Our research has been motivated by the task of forming a solution subgraph which satisfies given constraints. The problem is represented by an A=O graph. Our approach is to apply a suitably modified technique of dependencydirected backtracking. We present our formulation of the standard chronological backtracking algorithm in Prolog. Based on it, we have developed an enhanced algorithm which makes use of special heuristic knowledge. It involves also the technique of node marking. We have gathered experience with the prototype Prolog implementation of the algorithm in applying it to (one step of) the problem of building a software configuration. Our experience shows that Prolog programming techniques offer a considerable flexibility in implementing the above outlined tasks. Keywords. A=Ograph, nonchronological backtrack, Prolog 1 PROBLEM AREA AND GOAL Many problems to which artificial intelligence techniques are often applied can be described as constraint satisfaction problems. We...
Alamo: An Architecture for Integrating Heterogeneous Data Sources
 in Proceedings of the KRDB Workshop
, 1997
"... We are developing an architecture, Alamo, that addresses both the semantic and physical aspects of data integration. The Alamo architecture permits the interoperability of both data sources and semantic components. As a collection, the supported semantic components capture most basic forms of knowle ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
We are developing an architecture, Alamo, that addresses both the semantic and physical aspects of data integration. The Alamo architecture permits the interoperability of both data sources and semantic components. As a collection, the supported semantic components capture most basic forms of knowledge representation. Since the semantic integration of heterogeneous data sources requires some representation of the semantic content of the data source, the Alamo architecture forms an infrastructure for the development and possible integration of different forms of semantic integration of heterogeneous data sources. Central to the Alamo architecture is a CORBA compliant software bus called the Abstract Search Machine (ASM). The ASM augments a simple cursor class with methods that can be used to implement the marking, memoing and learning schemes exploited by clever search algorithms. The broad claim is that high performance implementations of se This research is partially funded by DAR...
BacktrackBounded Search in Polynomial Space
, 1994
"... We present and analyze a polynomially spacebounded backtrack algorithm for solving constraint satisfaction problems. We show the algorithm is capable of bounding worstcase runtime almost as effectively as the best exponential spaceconsuming schemes, and more effectively than various other sche ..."
Abstract
 Add to MetaCart
We present and analyze a polynomially spacebounded backtrack algorithm for solving constraint satisfaction problems. We show the algorithm is capable of bounding worstcase runtime almost as effectively as the best exponential spaceconsuming schemes, and more effectively than various other schemes including the cyclecutset method [Dechter, 90], pseudotree search [Freuder & Quinn, 85], and techniques exploiting nonseparable component decomposition of the constraint graph [Freuder, 85; Dechter, 87]. Experiments on randomly generated problems show the algorithm is capable of solving classes of problems on which a forward checking algorithm with dynamic search rearrangement [Haralick & Elliot, 80] often fails.