Logo do repositório
Tudo no RIUFPA
Documentos
Contato
Sobre
Ajuda
  • Português do Brasil
  • English
  • Español
  • Français
Entrar
Novo usuário? Clique aqui para cadastrar. Esqueceu sua senha?
  1. Início
  2. Pesquisar por Assunto

Navegando por Assunto "Entropia"

Filtrar resultados informando as primeiras letras
Agora exibindo 1 - 8 de 8
  • Resultados por página
  • Opções de Ordenação
  • Carregando...
    Imagem de Miniatura
    ItemAcesso aberto (Open Access)
    Entropy-based client selection strategy for federated learning over vehicular network environment
    (Universidade Federal do Pará, 2024-08-13) SOUSA, John Lucas Rodrigues Portilho de; ROSÁRIO, Denis Lima do; http://lattes.cnpq.br/8273198217435163; https://orcid.org/0000-0003-1119-2450; CERQUEIRA, Eduardo Coelho; http://lattes.cnpq.br/1028151705135221; https://orcid.org/0000-0003-2162-6523
    Federated Learning (FL) emerges as a promising solution to enable collaborative model training for autonomous vehicles while preserving privacy and addressing communication overhead issues. Efficient client selection for participation in the training process remains challenging, especially in scenarios with statistical heterogeneity of data distribution and client failure events. Client failure, an uncontrollable event during training, reduces accuracy, convergence, and speed. This master thesis introduces an entropy-based client selection mechanisms for FL over Vehicular Network environments with client failure and non-IID data distributions. The proposed method is compared to a random selection mechanism in both IID and non-IID scenarios, as well as scenarios with random client drops. The results demonstrate that entropy-based selection outperforms other methods regarding training loss, accuracy, and Area Under the Curve (AUC), particularly in high client dropout and non-IID scenarios. These findings highlight the importance of considering entropy data for client selection to address the challenges posed by client failure and statistical heterogeneity in FL over Vehicular Network.
  • Carregando...
    Imagem de Miniatura
    ItemAcesso aberto (Open Access)
    Estudo da magnetohidrodinâmica em dutos usando transformadas integrais
    (Universidade Federal do Pará, 2015-11-27) LOBATO, Benedito; SOUZA, Marcelo José Raiol; http://lattes.cnpq.br/2980566423137880; MACÊDO, Emanuel Negrão; http://lattes.cnpq.br/8718370108324505
    The present work presents the study of magnetohydrodinamics for the outflow of fluids in ducts with different geometric shapes. The integral transforms technics, in this case GITT, were used to obtain solutions of the momentum and energy transfer equations. In each case a hybrid (analytical-numerical) solution was obtained, and they were treated with computational codes that allowed the analysis of the velocity, temperature and entropy fields. The results were compared with the results obtained in classic situations of the literature and also with the actual results, showing the efficiency and capacity of the utilized technique.
  • Carregando...
    Imagem de Miniatura
    ItemAcesso aberto (Open Access)
    Inversão gravimétrica do relevo do embasamento usando regularização entrópica
    (Universidade Federal do Pará, 2007) OLIVEIRA, Alexandre de Souza; SILVA, João Batista Corrêa da; http://lattes.cnpq.br/1870725463184491
    We present a new gravity inversion method, which estimates the basement relief of a sedimentary basin, whose sediments may present a constant or a decreasing density contrast with depth relative to the basement. The method combines the minimization of the first-order entropy measure with the maximization of the zeroth-order entropy measure of the solution vector (depths to the basement). The interpretation model consists of a set of rectangular juxtaposed vertical prisms with known density contrasts and whose thicknesses represent the depths to the basement relief and are the parameters to be estimated. The minimization of the first-order entropy favors solutions presenting abrupt discontinuities, and the maximization of the zeroth-order entropy is employed just to prevent its excessive minimization. We applied our method to synthetic data simulating: (i) intracratonic basins with smooth or discontinuous relief, and (ii) faulted marginal basins. In the case of a smooth basement relief, it was well delineated by the entropic regularization, and by the global and weighted smoothness. In the case of a faulted basement, either in intracratonic or marginal basins, the entropic regularization and the weighted smoothness delineated the basement relief with good precision, differently from the global smoothness. The entropic regularization, however, did not require the knowledge of the basin’s maximum depth. We applied our method to four sets of real Bouguer anomalies. The first one comes from a profile across a bridge located in the campus of the Federal University of Pará, Belém. The solution delineated the known, discontinuous topography below the bridge. The second set comes from the northern portion of Steptoe Valley, Nevada. The solution delineated a discontinuous basement relief in accordance with the known geological setting of the area. The third set comes from the San Jacinto graben, situated in California, United States. The solution indicated an asymmetric graben, as already reported by other authors. The fourth data set comes from the Büyük Menderes graben, western Turkey. The solutions presented large vertical displacements on the northern border, which is confirmed by geological information.
  • Carregando...
    Imagem de Miniatura
    ItemAcesso aberto (Open Access)
    Inversão gravimétrica usando regularização entrópica
    (Universidade Federal do Pará, 2005) OLIVEIRA, Francisco de Souza; SILVA, João Batista Corrêa da; http://lattes.cnpq.br/1870725463184491
    We present a new gravity inversion method, which produces an apparent density contrast mapping on the horizontal plane by combining the minimization of the first-order entropy with the maximization of the zero-order entropy of the estimated density contrasts. The interpretation model consists of a grid of vertical, juxtaposed prisms in both horizontal directions. The top and the bottom of the gravity sources are assumed to be flat and horizontal, and the parameters to be estimated are the prism density contrasts. The maximization of the zero-order entropy is similar to the global smoothness constraint whereas the minimization of the first-order entropy favors solutions presenting sharp borders, so a judicious combination of both constrains may lead to solutions characterized by regions where the estimated density contrasts are virtually constant (in the case of homogeneous bodies), separated by sharp discontinuities. The method has been applied to synthetic data simulating the presence of intrusive bodies in sediments. The comparison of the results with those obtained with the global smoothness constraint method shows that both methods produce good and equivalent locations of the source positions, but the entropic regularization delineates the contour of the bodies with greater resolution, even in the case of 100 m wide bodies separated by a distance as small as 50 m. In the case that the depth to the top of the causative sources is neither flat nor horizontal, both method produced similar results. Both the proposed and the global smoothness methods have been applied to two sets of real data produced by igneous intrusions into metamorphic rocks. The first one is from the Matsitama region, northeast of Botswana. The application of both methods to this data set lead to similar results, indicating that depth to the sources top is neither flat nor horizontal. The second set is from the Cornuália region, England. The results produced by both methods in this case are reasonably different, indicating that the gravity source present an approximately flat and horizontal top. This is confirmed by the available geological information.
  • Carregando...
    Imagem de Miniatura
    ItemAcesso aberto (Open Access)
    Mapeamento de magnetização aparente usando regularização entrópica
    (Universidade Federal do Pará, 2007) VASCONCELOS, Suzan Sousa de; SILVA, João Batista Corrêa da; http://lattes.cnpq.br/1870725463184491
    We present a new apparent-magnetization mapping method on the horizontal plane that combines the minimization of the first-order entropy with the maximization of the zeroth-order entropy of the estimated magnetization contrasts. The interpretation model is a grid of vertical, juxtaposed prisms in both horizontal directions. We assume that the top and the bottom of the magnetic sources are flat and horizontal and estimate the prisms' magnetization contrasts. The minimization of the first-order entropy favors solutions presenting sharp borders, and the maximization of the zeroth-order entropy prevents the tendency of the source estimate to become a single prism. Thus, a judicious combination of both constraints may lead to solutions characterized by regions with virtually constant estimated magnetization contrasts separated by sharp discontinuities. We apply our method to synthetic data from simulated intrusive bodies in sediments that present flat and horizontal tops. By comparing our results with those obtained with the smoothness constraint, we show that both methods produce good and equivalent locations of the sources' central positions. However, the entropic regularization delineates the boundaries of the bodies with greater detail. Both the proposed and the smoothness constraints are applied to a real anomaly over a magnetic skarn in Butte Valley, Nevada, USA. The entropic regularization produced an estimated magnetization distribution with sharper boundaries, smaller volume, and higher apparent magnetization as compared with the results produced by the smoothness constraint.
  • Carregando...
    Imagem de Miniatura
    ItemAcesso aberto (Open Access)
    Mineração e entropia: notas sobre interação com processos de desenvolvimento
    (2007-06) MONTEIRO, Maurílio de Abreu; COELHO, Maria Célia Nunes
    This article presents theoretical and methodological assumptions that are relevant for the interpretation of the interactions between mining and regional development. It refutes the hypothesis of market mechanisms driving away the irreversibility of entropic processes, defended by the natural resources economy of neoclassical inspiration. The notion that the entropic degradation, from a social point of view, is inexorably negative – as proposed by the ecologic economy – is also refuted. Approaches of interpretation of the capitalist society as a system far away from balance are indicated as analytical alternatives. The author proposes that social repercussions of entropic degradation resulting from mining activities are connected to the interaction model between technical competence, historical, social, cultural and political features, and natural specialties and potentialities of a given social formation.
  • Carregando...
    Imagem de Miniatura
    ItemAcesso aberto (Open Access)
    Processos de relaxação em sistemas longe do equilíbrio termodinâmico
    (Universidade Estadual de Campinas, 1985-08-16) MACHADO, Carlos José Freire; VASCONCELLOS, Aurea Rosas; http://lattes.cnpq.br/0475932786473160
  • Carregando...
    Imagem de Miniatura
    ItemAcesso aberto (Open Access)
    Sistemas agroindustriais integrados: uma análise por meio da entropia de informação
    (2000-06) PONTE, Marcos Ximenes; VAN DYNE, Donald L.
    Using the mathematical theory of the information, this article analyzes four agroindustry systems of bioenergy production: the industry of alcohol from sugar-cane in Brazil, the industry of corn alcohol and the industry of biodiesel in United States, besides the industry of manioc flour from the Amazonian. Based on an important group of empiric information, evaluations of energy efficiency for different organization levels of the systems are made.
Logo do RepositórioLogo do Repositório
Nossas Redes:

DSpace software copyright © 2002-2025 LYRASIS

  • Configurações de Cookies
  • Política de Privacidade
  • Termos de Uso
  • Entre em Contato
Brasão UFPA