Wang, C.J. & Tsang, E.P.K., Solving constraint satisfaction problems using neural-networks, Proceedings, IEE Second International Conference on Artificial Neural Networks, 1991, 295-299  Home/Search   Document Details and Download   Summary   Related Articles   Check

....if C basic (v ( greedy select(v end while return no satisfying assignment found 2.4. 2 GENET GENET for constraint satisfaction problems is an approach for constraint satisfaction using a representation of a CSP as a neural net (see for example [44] developed by Wang and Tsang [7, 60, 62]. The method comes in two avors depending on if only binary or k ary constraints are to be allowed. GENET for binary CSP s In this model, each label hx; vi is represented by a node in the network, and each set of exactly two labels fhx; vi; hy; wig that are incompatible with respect to a ....

Wang, C., and Tsang, E. Solving constraint satisfaction problems using neural-networks. In IEE Second International Conference on Arti cial Neural Networks (1991).

....problems in practice, since if the network fails to converge, it can be stopped and started again. Unfortunately, the GDS network has shown to work effectively only on problems which have many solutions. It fails to solve problems which have few solutions or for which there are many local minima. Wang and Tsang [1991] have shown how to remedy this drawback. They have proposed a multilayer neural network and a learning rule that updates connection weights in order to escape from local optima. An issue closely related to constraint satisfaction is constraint relaxation. It has turned out that many practical ....

C. J. Wang and E. P. K. Tsang. Solving constraint satisfaction problems using neural networks. Department of Computer Science, University of Essex, 1991.

....recent interest in applying ANN to optimization and CSP. For examples, Hopfield network, the Boltzmann Machine and the elastic network have been used in the traveling salesman problems [15, 1, 10] the Tangram puzzles [19] and the N queens problem [25] with satisfactory results. Wang and Tsang [32, 28] propose GENET, a generic ANN model, for solving general CSP s with binary constraints. A massively parallel implementation of GENET may attain a theoretical speedup in the order of 10 8 over sequential heuristic search. Wang and Tsang [33] also propose a cascadable VLSI design for GENET. ANN, ....

....neural network has as answer. We say that the neural network model is probabilistically complete. 4. Incrementality) The model must be amenable to efficient incremental execution. Any ANN model that satisfies the above properties can be used in our framework. We have chosen the GENET model [32, 28] to demonstrate the feasibility of our proposal. In the following, we give a brief introduction to GENET and show how we adapt the execution model of GENET to an incremental version. At appropriate places, we explain why GENET satisfies the three criteria. 2.1 GENET GENET [32, 28] is a generic ....

[Article contains additional citation context not shown here]

C.J. Wang and E.P.K. Tsang. Solving constraint satisfaction problems using neural networks. In Proceedings of the IEE 2nd Conference on Artificial Neural Networks, pages 295--299, 1991.

....recent interest in applying ANN to optimization and CSP. For examples, Hopfield network, the Boltzmann Machine and the elastic network have been used in the traveling salesman problems [18, 1, 13] the Tangram puzzles [23] and the N queens problem [34] with satisfactory results. Wang and Tsang [43, 39] propose GENET, a generic ANN model, for solving general CSP s with binary constraints. Wang and Tsang [44] also propose a cascadable VLSI design for GENET. A VLSI implementation of GENET would provide a potential speed gain in an order of 10 6 to 10 8 over existing CSP languages running on ....

....ANN models are probabilistic in nature. 4. Incrementality) Adding new constraints to an existing solvable set of constraints is a frequent and primitive step in a CLP system. Thus the model must be amenable to efficient incremental execution. 2. 3 GENET: a Generic Neural Network Model GENET [43, 39] is a generic neural network simulator designed to solve binary CSP s with finite domains. It has also been applied to solve general CSP s such as the car sequencing problem [8] We describe the network structure and the convergence procedure of GENET. Then we explain the behavior of GENET s ....

C.J. Wang and E.P.K. Tsang. Solving constraint satisfaction problems using neural networks. In Proceedings of the IEE 2nd Conference on Artificial Neural Networks, pages 295--299, 1991.

....derivation, the issues dealt with in this paper do not arise. This paper is organised as follows. Section 2 briefly introduces some preliminaries for subsequent discussion. Section 3 describes how a stochastic solver can be used within a CLP system. In section 4, we briefly describe the GENET [2, 16, 19] model which is a probabilistic artificial neural networks (ANN) used as the constraint solver to demonstrate the feasibility of our approach. We show how we adopt the original model to support efficient incremental execution with backtracking on constraints. Section 5 describes our experimental ....

....deepening approach we can ensure that eventually we will find a success if one exists. Theorem 3.4: If P A j= 9 G then executing goal G using iterative deepening and a depth bounded usage strategy U will find a successful derivation. 4 A Constraint Solver : GENET We have chosen the GENET [2, 16, 19], a generic neural network simulator, to demonstrate the feasibility of our proposal. GENET can be used to solve general CSP s with finite domains. A GENET network consists of a cluster of nodes for each domain variable in a CSP. Each node denotes a value(label) in the corresponding domain ....

C.J. Wang and E.P.K. Tsang. Solving constraint satisfaction problems using neural networks. In Proceedings of the IEE 2nd Conference on Artificial Neural Networks, pages 295--299, 1991.

....increasing the successful derivation will be within the depth bound. By Theorem 4.4 this derivation has a non zero probability of being a successful derivation for U and ssolv. Execution continues until it becomes a successful derivation. 5 A Constraint Solver : GENET We have chosen the GENET [18, 20], a generic neural network simulator, to demonstrate the feasibility of our proposal. GENET can be used to solve binary CSP s with finite domains. A GENET network consists of a cluster of nodes for each domain variable in a CSP. Each node denotes a value(label) in the corresponding domain ....

C.J. Wang and E.P.K. Tsang. Solving constraint satisfaction problems using neural networks. In Proceedings of the IEE 2nd Conference on Artificial Neural Networks, 295--299, 1991.

....been recent interest in applying ANN to optimization and CSP. For examples, Hopfield network, the Boltzmann Machine and the elastic network have been used in the traveling salesman problems [12, 1, 8] the Tangram puzzles [15] and the N queens problem [21] with satisfactory results. Wang and Tsang [26, 22] proposed GENET, a generic ANN model, for solving general CSP s with binary constraints. A massively parallel implementation of GENET may attain a theoretical speedup in the order of 10 8 over sequential heuristic search. Wang and Tsang [27] also proposed a cascadable VLSI design for GENET. ....

....neural network has as answer. We say that the neural network model is probabilistically complete. 3. Incrementality) The model must be amenable to efficient incremental execution. Any ANN model that satisfies the above properties can be used in our framework. We have chosen the GENET model [26, 22] to demonstrate the feasibility of our proposal. In the following, we give a brief introduction to GENET and show how we adapt the execution model of GENET to an incremental version. Our statistics shows that the incremental version is at least as efficient as the batch version. At appropriate ....

[Article contains additional citation context not shown here]

C.J. Wang and E.P.K. Tsang. Solving constraint satisfaction problems using neural networks. In Proceedings of the IEE 2nd Conference on Artificial Neural Networks, pages 295--299, 1991.

.... [3] disjunctive scheduling [13] and firmware design [4] Stochastic search methods, such as simulated annealing, neural networks and evolutionary algorithms, have also had remarkable success in solving industrial CSP s [12] and CSP optimization problems [1, 5] Among the stochastic solvers, GENET [12, 14] is a neural network simulator to solve binary CSP s with finite domains. It is shown to be efficient for solving certain hard or large instances of CSP s [14, 2] GENET has been enhanced in a number of ways. It was augmented to handle atmost and illegal constraints in order to solve ....

....also had remarkable success in solving industrial CSP s [12] and CSP optimization problems [1, 5] Among the stochastic solvers, GENET [12, 14] is a neural network simulator to solve binary CSP s with finite domains. It is shown to be efficient for solving certain hard or large instances of CSP s [14, 2] . GENET has been enhanced in a number of ways. It was augmented to handle atmost and illegal constraints in order to solve car sequencing problems [2] Lee, Leung and Won [7] describe an extended form of GENET, E GENET, which has a generic representation scheme for handling general constraints. ....

[Article contains additional citation context not shown here]

C.J. Wang and E.P.K. Tsang. Solving constraint satisfaction problems using neural networks. In Proceedings of the IEE 2nd Conference on Artificial Neural Networks, 295--299, 1991.

....would be useful to others using GENET to solve large dicult constraint satisfaction problems and in particular problems with similar optimisation criteria. 3 Description of GENET GENET is based on a software simulation of a Neural Network. It was designed at Essex University and is introduced in [18] and [16] It is an iterative improvement technique using the principle of the min con icts heuristic [13] Tsang and Wang [18, 16] in 1991 and 1992 showed the success of GENET in solving binary CSPs. In 1994 Davenport and Tsang [5] enhanced GENET so it could handle some non binary constraints. ....

....optimisation criteria. 3 Description of GENET GENET is based on a software simulation of a Neural Network. It was designed at Essex University and is introduced in [18] and [16] It is an iterative improvement technique using the principle of the min con icts heuristic [13] Tsang and Wang [18, 16] in 1991 and 1992 showed the success of GENET in solving binary CSPs. In 1994 Davenport and Tsang [5] enhanced GENET so it could handle some non binary constraints. Lee et al. 12] have also proposed a system for adapting GENET for general non binary constraints, but the system developed by ....

[Article contains additional citation context not shown here]

C.J. Wang and E.P.K. Tsang. Solving constraint satisfaction problems using neural networks. In IEE Second International Conference on Articial Neural Networks, pages 295-299, 1991.

....criterion consists of removing a tabu classification from a move when the move leads to a solution better than that obtained so far. Another method that searches the space of complete labellings till the solution is found is based on connectionist approach represented by GENet al..gorithm [27]. The CSP problem is represented here as a network where the nodes correspond to values of all variables. The nodes representing values for one variable are grouped into a cluster and it is assumed that exactly one node in the cluster is switched on that means that respective value is chosen for ....

Wang, C,J., Tsang, E.P.K.: Solving constraint satisfaction problems using neural-networks, in: Proc. Second International Conference on Artificial Neural Networks, 1991

....criterion consists of removing a tabu classification from a move when the move leads to a solution better than that obtained so far. Another method that searches the space of complete labellings till the solution is found is based on connectionist approach represented by GENet al..gorithm [42]. The CSP problem is represented here as a network where the nodes correspond to values of all variables. The nodes representing values for one variable are grouped into a cluster and it is assumed that exactly one node in the cluster is switched on that means that respective value is chosen for ....

Wang, C,J., Tsang, E.P.K.: Solving constraint satisfaction problems using neural-networks, in: Proc. Second International Conference on Artificial Neural Networks, 1991

....than backtracking on certain tasks such as the n queens problems. Unfortunately, the GDS network has shown to work effectively only on problems which have many solutions. It fails to solve problems which have few solutions or for which there are many local minima. C. J. Wang and E. P. K. Tsang [149] have shown how to remedy this drawback. They have proposed a multi layer neural network and a learning rule that updates link weights in order to escape from local minima. An issue closely related to constraint satisfaction is constraint relaxation. It has turned out that many practical problems ....

C. J. Wang and E. P. K. Tsang. Solving constraint satisfaction problems using neural networks. Department of Computer Science, University of Essex, 1991.

....derivation, the issues dealt with in this paper do not arise. This paper is organised as follows. Section 2 briefly introduces some preliminaries for subsequent discussion. Section 3 describes how a stochastic solver can be used within a CLP system. In section 4, we briefly describe the GENET [2, 12, 14] model which is a probabilistic artificial neural networks(ANN) used as the constraint solver to demonstrate the feasibility of our approach. We show how we adopt the original model to support efficient incremental execution with backtracking on constraints. Section 5 describes our experimental ....

....deepening approach we can ensure that eventually we will find a success if one exists. Theorem 3.4: If P A j= 9 G then executing goal G using iterative deepening and a depthbounded usage strategy U will find a successful derivation. 4 A Constraint Solver : GENET We have chosen the GENET [2, 12, 14], a generic neural network simulator, to demonstrate the feasibility of our proposal. GENET can be used to solve binary CSP s with finite domains. A GENET network consists of a cluster of nodes for each domain variable in a CSP. Each node denotes a value(label) in the corresponding domain ....

C.J. Wang and E.P.K. Tsang. Solving constraint satisfaction problems using neural networks. In Proceedings of the IEE 2nd Conference on Artificial Neural Networks, pages 295--299, 1991.

....applying ANN to CSP s and constraint optimization problems. For examples, Hopfield network, the Boltzmann Machine and the elastic network have been used in the traveling salesman problems [18, 1, 13] the Tangram puzzles [23] and the N queens problem [34] with satisfactory results. Wang and Tsang [43, 39] propose GENET, a generic ANN model, for solving general CSP s with binary constraints. Wang and Tsang [44] also propose a cascadable VLSI design for GENET. A VLSI implementation of GENET would provide a potential speed gain in an order of 10 6 to 10 8 over existing CSP languages running on ....

....ANN models are probabilistic in nature. 4. Incrementality) Adding new constraints to an existing solvable set of constraints is a frequent and primitive step in a CLP system. Thus the model must be amenable to efficient incremental execution. 2. 3 GENET: a Generic Neural Network Model GENET [43, 39] is a generic neural network simulator designed to solve binary CSP s with finite domains. It has also been applied to solve general CSP s such as the car sequencing problem [8] We describe the network structure and the convergence procedure of GENET. Then we explain the behavior of GENET s ....

C.J. Wang and E.P.K. Tsang. Solving constraint satisfaction problems using neural networks. In Proceedings of the IEE 2nd Conference on Artificial Neural Networks, pages 295--299, 1991.

....been recent interest in applying ANN to optimization and CSP. For examples, Hopfield network, the Boltzmann Machine and the elastic network have been used in the traveling salesman problems [9, 1, 6] the Tangram puzzles [12] and the Nqueens problem [15] with satisfactory results. Wang and Tsang [19, 17] propose GENET, a generic ANN model, for solving general CSP s with binary constraints. A massively parallel implementation of GENET may attain a theoretical speedup in the order of 10 8 over sequential heuristic search. Wang and Tsang [20] also propose a cascadable VLSI design for GENET. ANN, ....

....nature of PROCLANN to the fullest extent. It means that if the CSP specified by the program is solvable, then there always exists a successful derivation and the answer network obtained has non zero probability to converge to any solution of the CSP. 4 GENET We have chosen the GENET model [19, 17] to demonstrate the feasibility of our proposal. GENET is a generic neural network simulator that can be used to solve general CSP s with finite domains. To illustrate how a GENET network is constructed for a CSP, let us take a simple but tight 2 binary CSP as example. Assume there are five ....

C.J. Wang and E.P.K. Tsang. Solving constraint satisfaction problems using neural networks. In Proceedings of the IEE 2nd Conference on Artificial Neural Networks, pages 295--299, 1991.

No context found.

Wang, C.J. & Tsang, E.P.K., Solving constraint satisfaction problems using neural-networks, Proceedings, IEE Second International Conference on Artificial Neural Networks, 1991, 295-299

....Like all other hill climbing algorithms, FLS suffers from the problem of settling in local optima. Guided local search (GLS) is a method for escaping local optima. GLS is built upon our experience in a connectionist method called GENET (it is a generalization of the GENET computation models) [27, 6, 26]. GLS is a algorithm for modifying local search algorithms. The basic idea is that costs and penalty values are associated to selected features of the candidate solutions. Selecting such features in an application is not difficult because the objective function is often made up of a number of ....

C.J. Wang and E.P.K. Tsang, "Solving constraint satisfaction problems using neural-networks", Proceedings of IEE Second International Conference on Artificial Neural Networks, 295-299 (1991).

No context found.

Wang, C.J. & Tsang, E.P.K., Solving constraint satisfaction problems using neuralnetworks, Proceedings of IEE Second International Conference on Artificial Neural Networks, 295-299, 1991

No context found.

Wang, C.J. & Tsang, E.P.K., Solving constraint satisfaction problems using neuralnetworks, Proceedings of IEE Second International Conference on Artificial Neural Networks, 295-299, 1991

No context found.

Wang, C.J. & Tsang, E.P.K., Solving constraint satisfaction problems using neuralnetworks, Proceedings of IEE Second International Conference on Artificial Neural Networks, 295-299, 1991

....such as the Travelling Salesman Problem (TSP) and the Quadratic Assignment Problem (QAP) have also been tackled with the method. GLS has been shown to be equally good if not better than the best TSP and QAP heuristic search algorithms [23, 21] GLS was derived from the GENET neural network [25] for Constraint Satisfaction Problems [19] and extends the approach used in GENET to the bulk of combinatorial optimisation problems. GLS belongs to a class of techniques known as Metaheuristics. Prominent members of this class include Tabu Search [8] Simulated Annealing [11] Genetic Algorithms ....

.... local search procedure for PCSPs which is also applicable to the RLFAP has been described in previous work of the authors [22] The scheme is based on the min conflicts heuristic of Minton et al. 13] for Constraint Satisfaction Problems and also the computational model of the GENET neural network [25, 6]. An 1 optimal type move is used which changes the value of one variable at a time. Starting from a random and complete assignment of values to variables, variables are examined in an arbitrary static order. Each time a variable is examined, the current value of the variable changes to the value ....

Wang, C.J., and Tsang, E., "Solving constraint satisfaction problems using neural-networks", Proceedings of IEE Second International Conference on Artificial Neural

....for NN AI, 1992 page 1 A CASCADABLE VLSI DESIGN FOR GENET Chang J. Wang and Edward P. K. Tsang Department of Computer Science University of Essex Wivenhoe Park Colchester CO4 3SQ United Kingdom INTRODUCTION This paper presents a VLSI design for a competitive neural network model, known as GENET (Wang and Tsang 1991), for solving Constraint Satisfaction Problems (CSP) The CSP is a mathematical abstraction of the problems in many AI application domains. In essence, a CSP can be defined as a triple (Z, D, C) where Z is a finite set of variables, D is a mapping from every variable to a domain, which is a ....

....can handle over constrained CSPs which are at the heart of many real life problems. To address these problems, a generic neural network approach, known as GENET, that effectively realizes a stochastic heuristic search, has been proposed to speed up the performance of CSP solvers (Tsang and Wang 1991). In the next section, we shall briefly describe the GENET model. This will be followed by a presentation of a cascadable VLSI design for implementing GENET. Proc. Int l Workshop on VLSI for NN AI, 1992 page 3 GENET GENET was inspired by the early attempt to apply neural network technique ....

[Article contains additional citation context not shown here]

Wang, C. J., & Tsang, E. T. K., "Solving constraint satisfaction problems using neural networks", Proceedings, IEE Second International Conference on Artificial Neural Networks, pp. 295-299, 1991.

....approach and the very encouraging results, the method could have an important contribution to the development of intelligent search techniques for combinatorial optimization. 1. Introduction Guided Local Search is the outcome of a research project with main aim to extend the GENET neural network [29,26,5] for constraint satisfaction problems to partial constraint satisfaction [6,26] and combinatorial optimization problems. Beginning with GENET, we developed a number of intermediate algorithms such as the Tunneling Algorithm [28] to conclude with Guided Local Search (GLS) presented in this paper. ....

C. J. Wang and E. Tsang, "Solving constraint satisfaction problems using neuralnetworks ", In Proceedings of IEE Second International Conference on Artificial Neural Networks, 295-299 (1991).

....settling in local minima, which prevents the problem solver from finding solutions when some exist. Our analysis and experiments have shown that the Heuristic Repair Method is only effective for CSPs for which there exist a large number of solutions, such as the N queens problem with large N [20]. In this paper, we describe a generic neural network approach for solving CSPs with binary constraints. The model that we propose is called GENET. The effectiveness of GENET is demonstrated by a simulator, which can generate connected networks dynamically for binary constraint problems, and ....

....structure. We have performed preliminary tests in applying GENET to networks constructed (by the experimenters) for instances of the car sequencing problem [8, 24] Like all the other tests so far, GENET finds solutions in those networks. Our approach to the car sequencing problem is outlined in [20]. Our long term research direction is to fabricate VLSI neuro chips for constructing CSP solvers for real life applications. 7 Concluding Summary We have pointed out that there are applications in which timely response by CSP solvers is so crucial that a limited degree of sacrifice in ....

Wang, C. J., & Tsang, E. T. K., "Solving constraint satisfaction problems using neural networks", Proc. IEE Second International Conference on Artificial Neural Networks, 1991

No context found.

C.J. Wang and E.P.K. Tsang. Solving constraint satisfaction problems using neural-networks. In Proceedings of IEE Second International Conference on Artificial Neural Networks, pages 295--299, 1991.

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