829 resultados para Algoritmos genéticos


Relevância:

100.00% 100.00%

Publicador:

Resumo:

Esse trabalho comparou, para condições macroeconômicas usuais, a eficiência do modelo de Redes Neurais Artificiais (RNAs) otimizadas por Algoritmos Genéticos (AGs) na precificação de opções de Dólar à Vista aos seguintes modelos de precificação convencionais: Black-Scholes, Garman-Kohlhagen, Árvores Trinomiais e Simulações de Monte Carlo. As informações utilizadas nesta análise, compreendidas entre janeiro de 1999 e novembro de 2006, foram disponibilizadas pela Bolsa de Mercadorias e Futuros (BM&F) e pelo Federal Reserve americano. As comparações e avaliações foram realizadas com o software MATLAB, versão 7.0, e suas respectivas caixas de ferramentas que ofereceram o ambiente e as ferramentas necessárias à implementação e customização dos modelos mencionados acima. As análises do custo do delta-hedging para cada modelo indicaram que, apesar de mais complexa, a utilização dos Algoritmos Genéticos exclusivamente para otimização direta (binária) dos pesos sinápticos das Redes Neurais não produziu resultados significativamente superiores aos modelos convencionais.

Relevância:

100.00% 100.00%

Publicador:

Resumo:

A área de pesquisa de testes não-destrutivos é muito importante, trabalhando com o diagnóstico e o monitoramento das condições dos componentes estruturais prevenindo falhas catastróficas. O uso de algoritmos genéticos para identificar mudanças na integridade estrutural através de mudanças nas respostas de vibração da estrutura é um método não-destrutivo que vem sendo pesquisado. Isto se deve ao fato de que são vantajosos em achar o mínimo global em situações difíceis de problemas de otimização, particularmente onde existem muitos mínimos locais como no caso de detecção de dano. Neste trabalho é proposto um algoritmo genético para localizar e avaliar os danos em membros estruturais usando o conceito de mudanças nas freqüências naturais da estrutura. Primeiramente foi realizada uma revisão das técnicas de detecção de dano das últimas décadas. A origem, os fundamentos, principais aspectos, principais características, operações e função objetivo dos algoritmos genéticos também são demonstrados. Uma investigação experimental em estruturas de materiais diferentes foi realizada a fim de se obter uma estrutura capaz de validar o método. Finalmente, se avalia o método com quatro exemplos de estruturas com danos simulados experimentalmente e numericamente. Quando comparados com técnicas clássicas de detecção dano, como sensibilidade modal, os algoritmos genéticos se mostraram mais eficientes. Foram obtidos melhores resultados na localização do que na avaliação das intensidades dos danos nos casos de danos propostos.

Relevância:

100.00% 100.00%

Publicador:

Resumo:

Este texto apresenta a tese de doutorado em Ciência da Computação na linha de pesquisa de Inteligência Artificial, dentro da área de IAD – Inteligência Artificial Distribuída (mais especificamente os Sistemas Multiagentes – SMA). O trabalho aborda a formação de grupos colaborativos em um ambiente multiagente interativo de aprendizagem na web, através da utilização de técnicas de Inteligência Artificial. O trabalho apresenta a definição e implementação de uma arquitetura de agentes modelados com algoritmos genéticos, integrada a um ambiente colaborativo de aprendizagem, o TelEduc. Inicialmente faz-se um breve estudo sobre as áreas envolvidas na tese: Informática na Educação, Educação a Distância, Inteligência Artificial, Inteligência Artificial Distribuída e Inteligência Artificial Aplicada à Educação. Abordam-se, também, as áreas de pesquisa que abrangem os Sistemas Multiagentes e os Algoritmos Genéticos. Após este estudo, apresenta-se um estudo comparativo entre ambientes de ensino e aprendizagem que utilizam a abordagem de agentes e a arquitetura proposta neste trabalho. Apresenta-se, também, a arquitetura de agentes proposta, integrada ao ambiente TelEduc, descrevendo-se o funcionamento de cada um dos agentes e a plataforma de desenvolvimento. Finalizando o trabalho, apresenta-se o foco principal do mesmo, a formação de grupos colaborativos, através da implementação e validação do agente forma grupo colaborativo. Este agente, implementado através de um algoritmo genético, permite a formação de grupos colaborativos seguindo os critérios estabelecidos pelo professor. A validação do trabalho foi realizada através de um estudo de caso, utilizando o agente implementado na formação de grupos colaborativos em quatro turmas de cursos superiores de Informática, na Região Metropolitana de Porto Alegre, em disciplinas que envolvem o ensino de programação de computadores.

Relevância:

100.00% 100.00%

Publicador:

Resumo:

Com o objetivo de estabelecer uma metodologia capaz segregar momentos de mercado e de identificar as características predominantes dos investidores atuantes em um determinado mercado financeiro, este trabalho emprega simulações geradas em um Mercado Financeiro Artificial baseado em agentes, utilizando um Algoritmo Genético para ajustar tais simulações ao histórico real observado. Para tanto, uma aplicação foi desenvolvida utilizando-se o mercado de contratos futuros de índice Bovespa. Esta metodologia poderia facilmente ser estendida a outros mercados financeiros através da simples parametrização do modelo. Sobre as bases estabelecidas por Toriumi et al. (2011), contribuições significativas foram atingidas, promovendo acréscimo de conhecimento acerca tanto do mercado alvo escolhido, como das técnicas de modelagem em Mercados Financeiros Artificiais e também da aplicação de Algoritmos Genéticos a mercados financeiros, resultando em experimentos e análises que sugerem a eficácia do método ora proposto.

Relevância:

100.00% 100.00%

Publicador:

Resumo:

A abordagem de Modelos Baseados em Agentes é utilizada para trabalhar problemas complexos, em que se busca obter resultados partindo da análise e construção de componentes e das interações entre si. Os resultados observados a partir das simulações são agregados da combinação entre ações e interferências que ocorrem no nível microscópico do modelo. Conduzindo, desta forma, a uma simulação do micro para o macro. Os mercados financeiros são sistemas perfeitos para o uso destes modelos por preencherem a todos os seus requisitos. Este trabalho implementa um Modelo de Mercado Financeiro Baseado em Agentes constituído por diversos agentes que interagem entre si através de um Núcleo de Negociação que atua com dois ativos e conta com o auxílio de formadores de mercado para promover a liquidez dos mercados, conforme se verifica em mercados reais. Para operação deste modelo, foram desenvolvidos dois tipos de agentes que administram, simultaneamente, carteiras com os dois ativos. O primeiro tipo usa o modelo de Markowitz, enquanto o segundo usa técnicas de análise de spread entre ativos. Outra contribuição deste modelo é a análise sobre o uso de função objetivo sobre os retornos dos ativos, no lugar das análises sobre os preços.

Relevância:

100.00% 100.00%

Publicador:

Resumo:

The predictive control technique has gotten, on the last years, greater number of adepts in reason of the easiness of adjustment of its parameters, of the exceeding of its concepts for multi-input/multi-output (MIMO) systems, of nonlinear models of processes could be linearised around a operating point, so can clearly be used in the controller, and mainly, as being the only methodology that can take into consideration, during the project of the controller, the limitations of the control signals and output of the process. The time varying weighting generalized predictive control (TGPC), studied in this work, is one more an alternative to the several existing predictive controls, characterizing itself as an modification of the generalized predictive control (GPC), where it is used a reference model, calculated in accordance with parameters of project previously established by the designer, and the application of a new function criterion, that when minimized offers the best parameters to the controller. It is used technique of the genetic algorithms to minimize of the function criterion proposed and searches to demonstrate the robustness of the TGPC through the application of performance, stability and robustness criterions. To compare achieves results of the TGPC controller, the GCP and proportional, integral and derivative (PID) controllers are used, where whole the techniques applied to stable, unstable and of non-minimum phase plants. The simulated examples become fulfilled with the use of MATLAB tool. It is verified that, the alterations implemented in TGPC, allow the evidence of the efficiency of this algorithm

Relevância:

100.00% 100.00%

Publicador:

Resumo:

The metaheuristics techiniques are known to solve optimization problems classified as NP-complete and are successful in obtaining good quality solutions. They use non-deterministic approaches to generate solutions that are close to the optimal, without the guarantee of finding the global optimum. Motivated by the difficulties in the resolution of these problems, this work proposes the development of parallel hybrid methods using the reinforcement learning, the metaheuristics GRASP and Genetic Algorithms. With the use of these techniques, we aim to contribute to improved efficiency in obtaining efficient solutions. In this case, instead of using the Q-learning algorithm by reinforcement learning, just as a technique for generating the initial solutions of metaheuristics, we use it in a cooperative and competitive approach with the Genetic Algorithm and GRASP, in an parallel implementation. In this context, was possible to verify that the implementations in this study showed satisfactory results, in both strategies, that is, in cooperation and competition between them and the cooperation and competition between groups. In some instances were found the global optimum, in others theses implementations reach close to it. In this sense was an analyze of the performance for this proposed approach was done and it shows a good performance on the requeriments that prove the efficiency and speedup (gain in speed with the parallel processing) of the implementations performed

Relevância:

100.00% 100.00%

Publicador:

Resumo:

In this work, the Markov chain will be the tool used in the modeling and analysis of convergence of the genetic algorithm, both the standard version as for the other versions that allows the genetic algorithm. In addition, we intend to compare the performance of the standard version with the fuzzy version, believing that this version gives the genetic algorithm a great ability to find a global optimum, own the global optimization algorithms. The choice of this algorithm is due to the fact that it has become, over the past thirty yares, one of the more importan tool used to find a solution of de optimization problem. This choice is due to its effectiveness in finding a good quality solution to the problem, considering that the knowledge of a good quality solution becomes acceptable given that there may not be another algorithm able to get the optimal solution for many of these problems. However, this algorithm can be set, taking into account, that it is not only dependent on how the problem is represented as but also some of the operators are defined, to the standard version of this, when the parameters are kept fixed, to their versions with variables parameters. Therefore to achieve good performance with the aforementioned algorithm is necessary that it has an adequate criterion in the choice of its parameters, especially the rate of mutation and crossover rate or even the size of the population. It is important to remember that those implementations in which parameters are kept fixed throughout the execution, the modeling algorithm by Markov chain results in a homogeneous chain and when it allows the variation of parameters during the execution, the Markov chain that models becomes be non - homogeneous. Therefore, in an attempt to improve the algorithm performance, few studies have tried to make the setting of the parameters through strategies that capture the intrinsic characteristics of the problem. These characteristics are extracted from the present state of execution, in order to identify and preserve a pattern related to a solution of good quality and at the same time that standard discarding of low quality. Strategies for feature extraction can either use precise techniques as fuzzy techniques, in the latter case being made through a fuzzy controller. A Markov chain is used for modeling and convergence analysis of the algorithm, both in its standard version as for the other. In order to evaluate the performance of a non-homogeneous algorithm tests will be applied to compare the standard fuzzy algorithm with the genetic algorithm, and the rate of change adjusted by a fuzzy controller. To do so, pick up optimization problems whose number of solutions varies exponentially with the number of variables

Relevância:

100.00% 100.00%

Publicador:

Resumo:

ln this work, it was deveIoped a parallel cooperative genetic algorithm with different evolution behaviors to train and to define architectures for MuItiIayer Perceptron neural networks. MuItiIayer Perceptron neural networks are very powerful tools and had their use extended vastIy due to their abiIity of providing great resuIts to a broad range of appIications. The combination of genetic algorithms and parallel processing can be very powerful when applied to the Iearning process of the neural network, as well as to the definition of its architecture since this procedure can be very slow, usually requiring a lot of computational time. AIso, research work combining and appIying evolutionary computation into the design of neural networks is very useful since most of the Iearning algorithms deveIoped to train neural networks only adjust their synaptic weights, not considering the design of the networks architecture. Furthermore, the use of cooperation in the genetic algorithm allows the interaction of different populations, avoiding local minima and helping in the search of a promising solution, acceIerating the evolutionary process. Finally, individuaIs and evolution behavior can be exclusive on each copy of the genetic algorithm running in each task enhancing the diversity of populations

Relevância:

100.00% 100.00%

Publicador:

Resumo:

The telecommunications industry has experienced recent changes, due to increasing quest for access to digital services for data, video and multimedia, especially using the mobile phone networks. Recently in Brazil, mobile operators are upgrading their networks to third generations systems (3G) providing to users broadband services such as video conferencing, Internet, digital TV and more. These new networks that provides mobility and high data rates has allowed the development of new market concepts. Currently the market is focused on the expansion of WiMAX technology, which is gaining increasingly the market for mobile voice and data. In Brazil, the commercial interest for this technology appears to the first award of licenses in the 3.5 GHz band. In February 2003 ANATEL held the 003/2002/SPV-ANATEL bidding, where it offered blocks of frequencies in the range of 3.5 GHz. The enterprises who purchased blocks of frequency were: Embratel, Brazil Telecom (Vant), Grupo Sinos, Neovia and WKVE, each one with operations spread in some regions of Brazil. For this and other wireless communications systems are implemented effectively, many efforts have been invested in attempts to developing simulation methods for coverage prediction that is close to reality as much as possible so that they may become believers and indispensable tools to design wireless communications systems. In this work wasm developed a genetic algorithm (GA's) that is able to optimize the models for predicting propagation loss at applicable frequency range of 3.5 GHz, thus enabling an estimate of the signal closer to reality to avoid significant errors in planning and implementation a system of wireless communication

Relevância:

100.00% 100.00%

Publicador:

Resumo:

Antenna arrays are able to provide high and controlled directivity, which are suitable for radiobase stations, radar systems, and point-to-point or satellite links. The optimization of an array design is usually a hard task because of the non-linear characteristic of multiobjective, requiring the application of numerical techniques, such as genetic algorithms. Therefore, in order to optimize the electronic control of the antenna array radiation pattem through genetic algorithms in real codification, it was developed a numerical tool which is able to positioning the array major lobe, reducing the side lobe levels, canceling interference signals in specific directions of arrival, and improving the antenna radiation performance. This was accomplished by using antenna theory concepts and optimization methods, mainly genetic algorithms ones, allowing to develop a numerical tool with creative genes codification and crossover rules, which is one of the most important contribution of this work. The efficiency of the developed genetic algorithm tool is tested and validated in several antenna and propagation applications. 11 was observed that the numerical results attend the specific requirements, showing the developed tool ability and capacity to handle the considered problems, as well as a great perspective for application in future works.

Relevância:

100.00% 100.00%

Publicador:

Resumo:

The pattern classification is one of the machine learning subareas that has the most outstanding. Among the various approaches to solve pattern classification problems, the Support Vector Machines (SVM) receive great emphasis, due to its ease of use and good generalization performance. The Least Squares formulation of SVM (LS-SVM) finds the solution by solving a set of linear equations instead of quadratic programming implemented in SVM. The LS-SVMs provide some free parameters that have to be correctly chosen to achieve satisfactory results in a given task. Despite the LS-SVMs having high performance, lots of tools have been developed to improve them, mainly the development of new classifying methods and the employment of ensembles, in other words, a combination of several classifiers. In this work, our proposal is to use an ensemble and a Genetic Algorithm (GA), search algorithm based on the evolution of species, to enhance the LSSVM classification. In the construction of this ensemble, we use a random selection of attributes of the original problem, which it splits the original problem into smaller ones where each classifier will act. So, we apply a genetic algorithm to find effective values of the LS-SVM parameters and also to find a weight vector, measuring the importance of each machine in the final classification. Finally, the final classification is obtained by a linear combination of the decision values of the LS-SVMs with the weight vector. We used several classification problems, taken as benchmarks to evaluate the performance of the algorithm and compared the results with other classifiers

Relevância:

100.00% 100.00%

Publicador:

Resumo:

Universidade Federal do Rio Grande do Norte