Figure 6 - uploaded by Stephen Beale
Content may be subject to copyright.
Source publication
This work 1 integrates three related AI search techniques -- constraint satisfaction, branch-and-bound and solution synthesis -- and applies the result to constraint satisfaction problems for which optimal answers are required. This method has already been shown to work well in natural language semantic analysis (Beale, et al, 1996); here we extend...
Similar publications
Citations
... A general¯exible solution method, based on dynamic programming (Bertele & Brioschi, 1972), which exploits graph decomposition is shown in Beale (1997). The constraint graph representing the whole problem is decomposed into subgraphs for which solutions are sought. ...
... The¯exible component may be implemented in a variety of ways, the Partial Constraint Satisfaction (Freuder & Wallace, 1992) and Fuzzy Constraint Satisfaction (Dubois et al., 1996a) frameworks are examples which oer a range of¯exible constraint types. In addition, Beale (1997) suggests an extension to the graph decomposition algorithm described in section 3.3 to support dynamic problems. The potential applications of DFCSP are wide, encompassing many dynamic problems that require more¯exibility than classical CSP can provide. ...
... Subgraph. Optimise with respect to assignments to x i (Adapted fromBeale (1997)) An evolving problem modelled via DCSP ...
Constraint satisfaction is a fundamental artificial intelligence technique offering a simple yet powerful representation. An increasing amount of attention has recently been paid to the development of constraint satisfaction techniques, and it has become clear that the original formulation of a static Constraint Satisfaction Problem (CSP) with hard, imperative constraints is insufficient to model many real problems. Two important extensions to the classical CSP framework which address some of these deficiencies are flexible and dynamic constraint satisfaction. This paper examines in detail classical, flexible and dynamic CSP. It reviews the motivations behind both extensions, and describes the techniques used to solve each type of problem. The paper employs a running example throughout to illustrate the ideas presented.
... The scheduler is split into two phases; each phase is modeled as a CSP. Figure 1 shows a graphical representation of the model. The input to the system are les that specify 1 The output of the system is an optimal valid timetable of courses, faculty assignments, and time slots. A detailed description of each of the phases follows. ...
This paper describes a scheduling system to fulfill the timetabling needs of the Computer Science program at Florida Institute of Technology. The system handles both general constraints of timetabling problems and constraints specific to our problem. It has the flexibility of allowing for either manual or automatic assignments. The scheduling process is split into two phases; each phase is modeled as a CSP. The first phase tackles resource allocation which is, in our case, the problem of assigning faculty to courses. The second phase tackles the problem of assigning consistent time slots to courses. Optimization techniques are used in both phases. The first phase optimizes faculty assignment to courses in terms of faculty preferences. The second phase optimizes time slot assignment in terms of differences with previous term. The main goal of the system is to provide high quality timetables to the Computer Science program with as little delay as possible. This goal was fulfilled as expe...
... 16. A neighborhood is a pair L P , N whereL P is a search space and N is a mapping N :L P −→ 2L P that defines set of reachable solutions N (s) ⊆L P from solution s. ...
La Optimización Combinatoria es una rama de la optimización en matemática aplicada y de la informática, relacionada con la investigación operativa, la teoría de algoritmos y la teoría de complejidad computacional, que se encuentra en la intersección de varios campos, tales como la inteligencia artificial, las matemáticas y la ingeniería del software. Los problemas de optimización combinatoria suelen consistir en encontrar valores para un conjunto de variables que están restringidas por un conjunto de restricciones, en algunos casos para optimizar una función dada (optimización) y en otros tan solo para encontrar una solución válida (satisfacción). Los algoritmos de optimización combinatoria resuelven instancias de problemas considerados difíciles en general gracias a una exploración inteligente del espacio de búsqueda, en parte reduciéndolo, en parte recorriéndolo de una forma eficiente. En esta tesis nos centramos en los algoritmos de optimización combinatoria que se consideran dentro del campo de la Inteligencia Artificial (aunque es cierto que la línea que lo separa del campo de la investigación operativa es muy fina), en vez de en algoritmos de investigación operativa. Así pues, métodos como la programación entera o el "Branch-and-Bound" no van a ser tratados. El objetivo de esta tesis es mostrar que diferentes técnicas pueden ser más adecuadas para diferentes problemas, y que técnicas híbridas que incluyen mecanismos de diferentes paradigmas se pueden beneficiar de las ventajas e intentar minimizar los inconvenientes de los mismos. Todo esto se muestra en esta tesis con la resolución de problemas difíciles de optimización combinatoria como completitud de cuasigrupos, golfista social, Golomb rulers, usando varias técnicas, que dan lugar al desarrollo de un algoritmo híbrido para encontrar Golomb rulers, que incorpora aspectos de algoritmos genéticos, búsqueda local, programación con restricciones e incluso cl Combinatorial Optimization is a branch of optimization in applied mathematics and computer science, related to operations research, algorithm theory and computational complexity theory that sits at the intersection of many fields, such as artificial intelligence, mathematics and software engineering. Combinatorial optimization problems commonly imply finding values to a set of variables which are restricted by a set of constraints, in some cases in order to optimize a certain function (optimization) and in others only to find a valid solution (satisfaction). Combinatorial optimization algorithms solve instances of problems that are believed to be hard in general by exploiting the usually large solution space of these instances. They can achieve this by reducing the effective size of the search space and by exploiting it effciently. In this thesis we focus on Combinatorial Optimization Algorithms which fall into the field of Artificial Intelligence (although the line that separates this field from Operations Research is very fine), instead of algorithms from the Operations Research field. Thus, methods such as Integer Programming (IP) or Branch and Bound (BB) are not considered. The goal of this thesis is to show that different approaches can be better suited for different problems, and that hybrid techniques which include mechanisms from different frameworks can benefit from their advantages while minimizing their drawbacks. All this is shown throughout this thesis by solving hard combinatorial optimization problems, such as quasigroup completion, social golfers, optimal Golomb rulers, using a variety of techniques, which lead to a hybrid algorithm for finding Golomb rulers that incorporates features of Genetic Algorithms, Local Search, Constraint Programming and even Clustering. Tesis doctoral leída en la Escuela Politécnica Superior de la Universidad Autónoma de Madrid el 4 de septiembre de 2006 La Optimización Combinatoria es una rama de la optimización en matemática aplicada y de la informática, relacionada con la investigación operativa, la teoría de algoritmos y la teoría de complejidad computacional, que se encuentra en la intersección de varios campos, tales como la inteligencia artificial, las matemáticas y la ingeniería del software. Los problemas de optimización combinatoria suelen consistir en encontrar valores para un conjunto de variables que están restringidas por un conjunto de restricciones, en algunos casos para optimizar una función dada (optimización) y en otros tan solo para encontrar una solución válida (satisfacción). Los algoritmos de optimización combinatoria resuelven instancias de problemas considerados difíciles en general gracias a una exploración inteligente del espacio de búsqueda, en parte reduciéndolo, en parte recorriéndolo de una forma eficiente. En esta tesis nos centramos en los algoritmos de optimización combinatoria que se consideran dentro del campo de la Inteligencia Artificial (aunque es cierto que la línea que lo separa del campo de la investigación operativa es muy fina), en vez de en algoritmos de investigación operativa. Así pues, métodos como la programación entera o el "Branch-and-Bound" no van a ser tratados. El objetivo de esta tesis es mostrar que diferentes técnicas pueden ser más adecuadas para diferentes problemas, y que técnicas híbridas que incluyen mecanismos de diferentes paradigmas se pueden beneficiar de las ventajas e intentar minimizar los inconvenientes de los mismos. Todo esto se muestra en esta tesis con la resolución de problemas difíciles de optimización combinatoria como completitud de cuasigrupos, golfista social, Golomb rulers, usando varias técnicas, que dan lugar al desarrollo de un algoritmo híbrido para encontrar Golomb rulers, que incorpora aspectos de algoritmos genéticos, búsqueda local, programación con restricciones e incluso cl Combinatorial Optimization is a branch of optimization in applied mathematics and computer science, related to operations research, algorithm theory and computational complexity theory that sits at the intersection of many fields, such as artificial intelligence, mathematics and software engineering. Combinatorial optimization problems commonly imply finding values to a set of variables which are restricted by a set of constraints, in some cases in order to optimize a certain function (optimization) and in others only to find a valid solution (satisfaction). Combinatorial optimization algorithms solve instances of problems that are believed to be hard in general by exploiting the usually large solution space of these instances. They can achieve this by reducing the effective size of the search space and by exploiting it effciently. In this thesis we focus on Combinatorial Optimization Algorithms which fall into the field of Artificial Intelligence (although the line that separates this field from Operations Research is very fine), instead of algorithms from the Operations Research field. Thus, methods such as Integer Programming (IP) or Branch and Bound (BB) are not considered. The goal of this thesis is to show that different approaches can be better suited for different problems, and that hybrid techniques which include mechanisms from different frameworks can benefit from their advantages while minimizing their drawbacks. All this is shown throughout this thesis by solving hard combinatorial optimization problems, such as quasigroup completion, social golfers, optimal Golomb rulers, using a variety of techniques, which lead to a hybrid algorithm for finding Golomb rulers that incorporates features of Genetic Algorithms, Local Search, Constraint Programming and even Clustering.
Given a set of radio broadcast programs, the radio broadcast scheduling problem is to allocate a set of devices to transmit the programs to achieve the optimal sound quality. In this article, we propose a complete algorithm to solve the problem, which is based on a branch-and-bound (BnB) algorithm. We formulate the problem with a new model, called constrained maximum weighted bipartite matching (CMBM), i.e., the maximum matching problem on a weighted bipartite graph with constraints. For the reduced matching problem, we propose a novel BnB algorithm by introducing three new strategies, including the highest quality first, the least conflict first and the more edge first. We also establish an upper bound estimating function for pruning the search space of the algorithm. The experimental results show that our new algorithm can quickly find the optimal solution for the radio broadcast scheduling problem at small scales, and has higher scalability for the problems at large scales than the existing complete algorithm.
This paper describes ongoing work in carrying out the semantic analysis of texts and reference resolution in a control structure that permits each process to inform the other, rather than in a more traditional, unidirectional fashion (semantics followed by reference resolution). We concentrate on situations in which a polysemous predicate cannot be lexically disambiguated until the meaning of one of its arguments has been specified, and that can only be accomplished with the help of reference resolution procedures. As a sidebar, we briefly introduce our “feature value bundling” approach to configuring reference resolution engines without the need for large annotated corpora.
