Catálogo de publicaciones - libros

Update of knowledge bases is becoming an important topic in Artificial Intelligence and a key problem in knowledge representation and reasoning. One of the latest ideas to update logic programs is choosing between models of Minimal Generalised Answer Sets to overcome disadvantages of previous approaches. This paper describes an implementation of the declarative version of updates sequences that has been proposed as an alternative to syntax-based semantics. One of the main contributions of this implementation is to use DLV’s Weak Constraints to compute the model(s) of an update sequence, besides presenting the precise definitions proposed by the authors and an online solver. As a result, the paper makes an outline of the basic structure of the system, describes the employed technology, discusses the major process of computing the models, and illustrates the system through examples.

Hardware implementation of fuzzy systems plays important role in many industrial applications of fuzzy logic. The most popular applications of fuzzy hardware systems were found in the domain of control systems but the area of application of these systems is extending on other areas such as signal processing, pattern recognition, expert systems etc. The digital fuzzy hardware systems usually use only basic operations of fuzzy logic like min, max and some others, first, due to their popularity in traditional fuzzy control systems and, second, due to the difficulties of hardware implementation of more complicated operations, e.g. parametric classes of -norms and -conorms. But for extending the area of applications and flexibility of fuzzy hardware systems it is necessary to develop the methods of digital hardware implementation of wide range of fuzzy operations. The paper studies the problem of digital hardware implementation of fuzzy parametric conjunction and disjunction operations. A new class of such operations is proposed which is simple for digital hardware implementation and is flexible, due to its parametric form, for possible tuning in fuzzy models. The methods of hardware implementation of these operations in digital systems are proposed.

In this paper, we present a novel model of an artificial immune system (AIS), based on the process that suffers the T-Cell. The proposed model is used for solving constrained (numerical) optimization problems. The model operates on three populations: Virgins, Effectors and Memory. Each of them has a different role. Also, the model dynamically adapts the tolerance factor in order to improve the exploration capabilities of the algorithm. We also develop a new mutation operator which incorporates knowledge of the problem. We validate our proposed approach with a set of test functions taken from the specialized literature and we compare our results with respect to Stochastic Ranking (which is an approach representative of the state-of-the-art in the area) and with respect to an AIS previously proposed.

In this work we design a genetic representation and its genetic operators to encode individuals for evolving Dynamic System Models in a Qualitative Differential Equation form, for System Identification. The representation proposed, can be implemented in almost every programming language without the need of complex data structures, this representation gives us the possibility to encode an individual whose phenotype is a Qualitative Differential Equation in QSIM representation. The Evolutionary Computation paradigm we propose for evolving structures like those found in the QSIM representation, is a variation of Genetic Programming called Gene Expression Programming. Our proposal represents an important variation in the multi-gene chromosome structure of Gene Expression Programming at the level of the gene codification structure. This gives us an efficient way of evolving QSIM Qualitative Differential Equations and the basis of an Evolutionary Computation approach to Qualitative System Identification.

This paper presents a particle swarm optimizer for solving constrained optimization problems which adopts a very small population size (five particles). The proposed approach uses a reinitialization process for preserving diversity, and does not use a penalty function nor it requires feasible solutions in the initial population. The leader selection scheme adopted is based on the distance of a solution to the feasible region. In addition, a mutation operator is incorporated to improve the exploratory capabilities of the algorithm. The approach is tested with a well-know benchmark commonly adopted to validate constraint-handling approaches for evolutionary algorithms. The results show that the proposed algorithm is competitive with respect to state-of-the-art constraint-handling techniques. The number of fitness function evaluations that the proposed approach requires is almost the same (or lower) than the number required by the techniques of the state-of-the-art in the area.

Flock traffic navigation based on negotiation (FTN) is a new approach for solving traffic congestion problems in big cities. Early works suppose a navigation path based on a bone-structure made by initial, ending and geometrical intersection points of two agents and their rational negotiations. In this paper we present original methods based on clustering analysis to allow other agents to enter or abandon flocks according to their own self interests.

A new numerical optimization algorithm inspired by political competitions during parliamentary head elections is proposed in this paper. Competitive behaviors could be observed in many aspects of human social life. Our proposed algorithm is a stochastic, iterative and population-based global optimization technique like genetic algorithms and particle swarm optimizations. Particularly, our method tries to simulate the intra and inter group competitions in trying to take the control of the parliament. Performance of this method for function optimization over some benchmark multi-dimensional functions, of which global and local minimums are known, is compared with traditional genetic algorithms.

We have developed an algorithm using a Design of Experiments technique for reduction of search-space in global optimization problems. Our approach is called Domain Optimization Algorithm. This approach can efficiently eliminate search-space regions with low probability of containing a global optimum. The Domain Optimization Algorithm approach is based on eliminating non-promising search-space regions, which are identifyed using simple models (linear) fitted to the data. Then, we run a global optimization algorithm starting its population inside the promising region. The proposed approach with this heuristic criterion of population initialization has shown relevant results for tests using hard benchmark functions.

The (TTP) is an optimization problem that represents some types of sports timetabling, where the objective is to minimize the total distance traveled by the teams. This work proposes the use of a hybrid heuristic to solve the mirrored TTP (mTTP), called (*CS), that consists in detecting supposed promising search areas based on clustering. The validation of the results will be done in benchmark problems available in literature and real benchmark problems, e.g. Brazilian Soccer Championship.

A new stochastic procedure is applied to optimization problems that arise in the nonlinear modeling of data. The proposed technique is an implementation of a recently introduced algorithm for the construction of probability densities that are consistent with the asymptotic statistical properties of general stochastic search processes. The obtained densities can be used, for instance, to draw suitable starting points in nonlinear optimization algorithms. The proposed setup is tested on a benchmark global optimization example and in the weight optimization of an artificial neural network model. Two additional examples that illustrate aspects that are specific to data modeling are outlined.