Notre objectif dans cette thèse est de résoudre un problème quadratique binaire sous contraintes mixtes par les techniques d'optimisation DC. Puisque l'optimisation DC a prouvé son efficacité pour résoudre des problèmes de grandes tailles dans différents domaines, nous avons décidé d'appliquer cette approche d'optimisation pour résoudre ce problème. La partie la plus importante de l'optimisation DC est le choix d'une décomposition adéquate qui facilite la détermination et accélère la convergence de deux suites construites. La première suite converge vers la solution optimale du problème primal et la seconde converge vers la solution optimale du problème dual. Dans cette thèse, nous proposons deux décompositions DC efficaces et simples à manipuler. L'application de l'algorithme DC (DCA) nous conduit à résoudre à chaque itération un problème quadratique convexe avec des contraintes mixtes, linéaires et quadratiques. Pour cela, il faut trouver une méthode efficace et rapide pour résoudre ce dernier problème à chaque itération. Pour cela, nous appliquons trois méthodes différentes: la méthode de Newton, la programmation semi-définie positive et la méthode de points intérieurs. Nous présentons les résultats numériques comparatifs sur les mêmes repères de ces trois approches pour justifier notre choix de la méthode la plus rapide pour résoudre efficacement ce problème. / Our objective in this work is to solve a binary quadratic problem under mixed constraints by the techniques of DC optimization. As DC optimization has proved its efficiency to solve large-scale problems in different domains, we decided to apply this optimization approach to solve this problem. The most important part of D.C. optimization is the choice of an adequate decomposition that facilitates determination and speeds convergence of two constructed suites where the first converges to the optimal solution of the primal problem and the second converges to the optimal solution of the dual problem. In this work, we propose two efficient decompositions and simple to manipulate. The application of the DC Algorithm (DCA) leads us to solve at each iteration a convex quadratic problem with mixed, linear and quadratic constraints. For it, we must find an efficient and fast method to solve this last problem at each iteration. To do this, we apply three different methods: the Newton method, the semidefinite programing and interior point method. We present the comparative numerical results on the same benchmarks of these three approaches to justify our choice of the fastest method to effectively solve this problem.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-03-22T13:26:29Z No. of bitstreams: 1 fernandacampanharofavoreto.pdf: 5158420 bytes, checksum: b2d8536acc7b21e77d47e123d077e434 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-04-24T02:21:33Z (GMT) No. of bitstreams: 1 fernandacampanharofavoreto.pdf: 5158420 bytes, checksum: b2d8536acc7b21e77d47e123d077e434 (MD5) / Made available in DSpace on 2016-04-24T02:21:33Z (GMT). No. of bitstreams: 1 fernandacampanharofavoreto.pdf: 5158420 bytes, checksum: b2d8536acc7b21e77d47e123d077e434 (MD5) Previous issue date: 2013-05-17 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Bromeliaceae é um caso de excepcional radiação adaptativa entre as angiospermas, e posssui na costa leste do Brasil um de seus centros de diversidade e endemismo, destacando-se também pela sua importância ecológica e econômica. Estes fatores têm colocado esse grupo em evidência no contexto do manejo e conservação de plantas, e a obtenção de informações sobre a biologia dos organismos é etapa fundamental para o estabelecimento de estratégias de conservação. O Corredor Ecológico Prioritário Burarama-Pacotuba-Cafundó (CEPBPC), localizado em Cachoeiro de Itapemirim- Espírito Santo, foi instituído no âmbito do Corredor Central da Mata Atlântica e considerado área prioritária para conservação, porém está presente em uma região cuja composição e similaridade da flora não possuem um conhecimento consolidado. Este estudo visa ampliar o conhecimento sobre a diversidade taxonomica e conservação de Bromeliaceae em fragmentos de Floresta do estado do Espírito Santo. Para isso, um levantamento florístico da família e a caracterização qualitativa da distribuição das espécies nos diferentes habitats foram conduzidos no CEPBPC. Aspectos da similaridade floristica entre a área de estudo e outras localidades de Floresta Atlântica também foram investigados por análise multivariada de classificação e ordenação, e análise da correlação com distância geográfica. Foram levantadas 51 species distribuídas em 17 gêneros no CEPBPC, sendo Vriesea e Tillandsia, os gêneros com maior riqueza de especies. Foram ainda registradas nove espécies ameaçadas de extinção e a primeira ocorrência de V. correiaaraujoi Pereira & Penna para o estado do Espírito Santo. Além dos ambientes de floresta, a região de Floresta Estacional Semidecidual e com altitude inferior a 300m apresentaram maior riqueza. Porém destacam-se pela concentração de species e necessidade de políticas de conservação, as fisionomias de Floresta Ombrófila, os afloramentos rochosos presentes na região de Florestas Estacionais Semideciduais e, entre os ambientes antropizados, as áreas associadas a rochas, alagáveis ou ripárias.Na análise de classificação, o CEPBPC mostrou maior similaridade florística com a região da Pedra dos Pontões de Mimoso do Sul e posteriormente com áreas da região serrana do Espírito Santo, formando um grande bloco com regiões costeiras do estado e extremo sul da Bahia. Nas análises de ordenação, o CEPBPC ocupou uma posição intermediária ao longo do gradiente latitudinal da Floresta Atlântica, mostrando-se próximo também de áreas localizadas no norte do estado do Rio de Janeiro. Os resultados de ambas as análises mostraram-se complementares para explicar as relações florísticas de Bromeliaceae da região, e a influência da distância geográfica sobre estes padrões foi confirmada pela análise de correlação. No entanto, estudos adicionais envolvendo análise de direta de variáveis geoclimáticas podem acrescentar novas informações para elucidar os fatores que determinam os padrões de similaridade florística de Bromeliaceae na região. / Bromeliaceae is an excepcional case of adaptive radiation among angiosperms, and has in the east coast of Brazil one of its centers of diversity and Endemism. Standing out also for its ecologic and economic importance. These factors have placed this group in evidence in the context of the management and conservation of plants, so obtaining information on the biology of these organisms is a crucial step for the establishment of conservation strategies. The “Priority Ecological Corridor Burarama-Pacotuba-Cafundó” (CEPBPC), located in Cachoeiro de Itapemirim in the state of Espírito Santo, was established under the “Central Corridor of the Atlantic Forest” and is considered a priority area for Conservation, but it is present in an area in which the flora’s composition and similarity does not have a consolidated knowledge. The present study aims to increase knowledge of the taxonomic diversity and conservation of Bromeliaceae fragments of Atlantic Forest in the state of Espírito Santo. For that purpose, a floristic inventory of family and the qualitative characterization of the species’ distribution in different habitats were conducted in CEPBPC. Aspects of floristic similarity among the study area and other places of Atlantic Forest were also investigated by multivariate analysis, classification and ordination, and analysis of correlation with geographic distance. Fifty one species distributed in seventeen genera in CEPBPC have been raised, being Vriesea and Tillandsia, the genera with the richest of species. Nine species threatened to extinction and the first record of distribuition of V. correia-araujoi Pereira & Penna in the state of Espírito Santo have also been recorded. Besides the forest environments, the region of Semideciduous Seasonal Forest and with less than 300m altitude had higher affluence of species.However, the species that stand out for their concentration and the need for conservation policies, the physiognomies of Ombrophilous Forest, the rocky outcrops in the region of Semideciduous Seasonal Forests and among anthropogenic environments, the areas associated with rocks, or floodable Riparian. In the classification analysis, the CEPBPC showed greater floristic similarity with the region of Pedra dos Pontões in Mimoso do Sul and later with areas in the mountainous region in the state of Espírito Santo, forming a large block with coastal regions of this state and southern Bahia. In ordination analysis, the CEPBPC held an intermediate position along the latitudinal gradient of Atlantic Forest, being also close to areas located in the northern state of Rio de Janeiro. The results of both analyzes proved to be complementary to explain the floristic relationships of Bromeliaceae in the region, and the influence of the geographic distance on these patterns confirmed by correlation analysis. However, additional studies involving direct analysis of geoclimatic variables may add new information to elucidate the factors that determine the patterns of floristic similarity of Bromeliaceae in the region.

La classification (supervisée, non supervisée et semi-supervisée) est une thématique importante de la fouille de données. Dans cette thèse, nous nous concentrons sur le développement d'approches d'optimisation pour résoudre certains types des problèmes issus de la classification de données. Premièrement, nous avons examiné et développé des algorithmes pour résoudre deux problèmes classiques en apprentissage non supervisée : la maximisation du critère de modularité pour la détection de communautés dans des réseaux complexes et les cartes auto-organisatrices. Deuxièmement, pour l'apprentissage semi-supervisée, nous proposons des algorithmes efficaces pour le problème de sélection de variables en semi-supervisée Machines à vecteurs de support. Finalement, dans la dernière partie de la thèse, nous considérons le problème de sélection de variables en Machines à vecteurs de support multi-classes. Tous ces problèmes d'optimisation sont non convexe de très grande dimension en pratique. Les méthodes que nous proposons sont basées sur les programmations DC (Difference of Convex functions) et DCA (DC Algorithms) étant reconnues comme des outils puissants d'optimisation. Les problèmes évoqués ont été reformulés comme des problèmes DC, afin de les résoudre par DCA. En outre, compte tenu de la structure des problèmes considérés, nous proposons différentes décompositions DC ainsi que différentes stratégies d'initialisation pour résoudre un même problème. Tous les algorithmes proposés ont été testés sur des jeux de données réelles en biologie, réseaux sociaux et sécurité informatique / Classification (supervised, unsupervised and semi-supervised) is one of important research topics of data mining which has many applications in various fields. In this thesis, we focus on developing optimization approaches for solving some classes of optimization problems in data classification. Firstly, for unsupervised learning, we considered and developed the algorithms for two well-known problems: the modularity maximization for community detection in complex networks and the data visualization problem with Self-Organizing Maps. Secondly, for semi-supervised learning, we investigated the effective algorithms to solve the feature selection problem in semi-supervised Support Vector Machine. Finally, for supervised learning, we are interested in the feature selection problem in multi-class Support Vector Machine. All of these problems are large-scale non-convex optimization problems. Our methods are based on DC Programming and DCA which are well-known as powerful tools in optimization. The considered problems were reformulated as the DC programs and then the DCA was used to obtain the solution. Also, taking into account the structure of considered problems, we can provide appropriate DC decompositions and the relevant choice strategy of initial points for DCA in order to improve its efficiency. All these proposed algorithms have been tested on the real-world datasets including biology, social networks and computer security

L’image est une des informations les plus importantes dans la vie. Avec le développement rapide des dispositifs d’acquisition d’images numériques par exemple les appareils photo numériques, les caméras de téléphones, les appareils d’imagerie médicale ou les dispositifs d’imagerie satellite..., les besoins de traitement et d’analyse des images sont de plus en plus croissants. Ils concernent les problèmes de l’acquisition, du stockage des images, de l’amélioration ou de l’information d’extraction d’une image,... Dans cette thèse, nous étudions le traitement et l’analyse des problèmes: acquisition comprimée, apprentissage de dictionnaire et débruitage d’images, segmentation d’images. La méthode que nous décrivons se base sur les approches d’optimisation déterministe, nommées la programmation DC (Difference of Convex functions) et DCA (Difference of Convex Algorithms), pour la résolution des problèmes d’analyse d’images cités précédemment. 1. Acquisition comprimée: une technique de traitement du signal pour acquérir et reconstruire un signal respectant les limites traditionnelles du théorème d’échantillonnage de Nyquist–Shannon, en trouvant la solution la plus parcimonieuse d’un système linéaire sous-déterminé. Cette méthode apporte la parcimonie ou la compressibilité du signal lorsqu’il est représenté dans une base ou un dictionnaire approprié qui permet au signal entier d’être déterminé à partir de certains mesures relatives. Dans cette thématique, nous nous intéressons à deux problèmes. Le premier est de trouver la représentation parcimonieuse d’un signal. Le second est la récupération du signal à partir de ses mesures compressées sur une base incohérente ou un dictionnaire. Les deux problèmes ci-dessus conduisent à résoudre un problème d’optimisation non convexe. Nous étudions trois modèles avec quatre approximations pour ces problèmes. Des algorithmes appropriés basés sur la programmation DC et DCA sont présentés. 2. Apprentissage du dictionnaire: Nous avons vu la puissance et les avantages de la représentation parcimonieuse des signaux dans l’acquisition comprimée. La représentation parcimonieuse d’un signal entier dépend non seulement des algorithmes de représentation mais aussi de la base ou du dictionnaire qui sont utilisés dans la représentation. Ainsi conduit un problème critique et les autres applications d’une manière naturelle. Au lieu d’utiliser une base fixe, comme wavelets (ondelettes) ou Fourier, on peut apprendre un dictionnaire, la matrice D, pour optimiser la représentation parcimonieuse d’une large classe de signaux donnés. La matrice D est appelée le dictionnaire appris. Pour ce problème, nous avons proposé un algorithme efficace basé sur DCA qui comprend deux étapes: la première étape - codage parcimonieux; le seconde étape - dictionnaire mis à jour. Une application de ce problème, débruitage d’images, est également considérée. 3. Segmentation d’images: il s’agit de partitionner une image numérique en segments multiples (ensembles des pixels). Le but de la segmentation est de simplifier et/ou de modifier la représentation d’une image en une forme qui est plus significative et plus facile à analyser. Nous avons développé une méthode efficace pour la segmentation d’images via le clustering flou avec la pondération de variables. Nous étudions également une application médicale qui est le problème de comptage de cellules. Nous proposons une combinaison de phase de segmentation et des opérations morphologiques pour compter automatiquement le nombre de cellules. Notre approche donne des résultats prometteurs dans la comparaison avec l’analyse manuelle traditionnelle en dépit de la densité cellulaire très élevée / Image is one of the most important information in our lives. Along with the rapid development of digital image acquisition devices such as digital cameras, phone cameras, the medical imaging devices or the satellite imaging devices..., the needs of processing and analyzing images is more and more demanding. It concerns with the problem of image acquiring, storing, enhancing or extracting information from an image,... In this thesis, we are considering the image processing and analyzing problems including: compressed sensing, dictionary learning and image denoising, and image segmentation. Our method is based on deterministic optimization approach, named the DC (Difference of Convex) programming and DCA (Difference of Convex Algorithms) for solving some classes of image analysis addressed above. 1. Compressed sensing is a signal processing technique for efficiently acquiring and reconstructing a signal, which is breaking the traditional limits of sampling theory of Nyquist–Shannon by finding solutions to underdetermined linear systems. This takes advantage of the signal’s sparseness or compressibility when it is represented in a suitable basis or dictionary, which allows the entire signal to be determined from few relative measurements. In this problem, we are interested in two aspects phases. The first one is finding the sparse representation of a signal. The other one is recovering the signal from its compressed measurements on an incoherent basis or dictionary. These problems lead to solve a NP–hard nonconvex optimization problem. We investigated three models with four approximations for each model. Appropriate algorithms based on DC programming and DCA are presented. 2. Dictionary learning: we have seen the power and the advantages of the sparse representation of signals in compressed sensing. Finding out the sparsest representation of a set of signals depends not only on the sparse representation algorithms but also on the basis or the dictionary used to represent them. This leads to the critical problems and other applications in a natural way. Instead of using a fixed basis such as wavelets or Fourier, one can learn the dictionary, a matrix D, to optimize the sparsity of the representation for a large class of given signals (data). The matrix D is called the learned dictionary. For this problem, we proposed an efficient DCA based algorithm including two stages: sparse coding and dictionary updating. An application of this problem, image denoising, is also considered. 3. Image segmentation: partitioning a digital image into multiple segments (sets of pixels). The goal of segmentation is to simplify and/or change the representation of an image into a form that is more meaningful and easier to analyze. We have developed an efficient method for image segmentation via feature weighted fuzzy clustering model. We also study an application of image segmentation for cell counting problem in medicine. We propose a combination of segmentation phase and morphological operations to automatically count the number of cells. Our approach gives promising results in comparison with the traditional manual analysis in despite of the very high cell density

Submitted by Andréia Arantes Felício null (andreia_a.felicio@hotmail.com) on 2017-03-27T17:43:13Z No. of bitstreams: 1 Tese Andréia Arantes Felício.pdf: 4621912 bytes, checksum: fcb824edfc225a271b66a12fc039e014 (MD5) / Rejected by Luiz Galeffi (luizgaleffi@gmail.com), reason: Solicitamos que realize uma nova submissão seguindo as orientações abaixo: No campo “Versão a ser disponibilizada online imediatamente” foi informado que seria disponibilizado o texto completo porém no campo “Data para a disponibilização do texto completo” foi informado que o texto completo deverá ser disponibilizado apenas 6 meses após a defesa. Caso opte pela disponibilização do texto completo apenas 6 meses após a defesa selecione no campo “Versão a ser disponibilizada online imediatamente” a opção “Texto parcial”. Esta opção é utilizada caso você tenha planos de publicar seu trabalho em periódicos científicos ou em formato de livro, por exemplo e fará com que apenas as páginas pré-textuais, introdução, considerações e referências sejam disponibilizadas. Se optar por disponibilizar o texto completo de seu trabalho imediatamente selecione no campo “Data para a disponibilização do texto completo” a opção “Não se aplica (texto completo)”. Isso fará com que seu trabalho seja disponibilizado na íntegra no Repositório Institucional UNESP. Por favor, corrija esta informação realizando uma nova submissão. Agradecemos a compreensão. on 2017-03-29T19:14:18Z (GMT) / Submitted by Andréia Arantes Felício null (andreia_a.felicio@hotmail.com) on 2017-03-29T19:20:52Z No. of bitstreams: 1 Tese Andréia Arantes Felício.pdf: 4621912 bytes, checksum: fcb824edfc225a271b66a12fc039e014 (MD5) / Approved for entry into archive by Luiz Galeffi (luizgaleffi@gmail.com) on 2017-03-29T19:24:57Z (GMT) No. of bitstreams: 1 felicio_aa_dr_sjrp.pdf: 4621912 bytes, checksum: fcb824edfc225a271b66a12fc039e014 (MD5) / Made available in DSpace on 2017-03-29T19:24:57Z (GMT). No. of bitstreams: 1 felicio_aa_dr_sjrp.pdf: 4621912 bytes, checksum: fcb824edfc225a271b66a12fc039e014 (MD5) Previous issue date: 2017-02-21 / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / O Brasil é o principal produtor de cana-de-açúcar do mundo e, para que esta demanda seja suprida, muitos compostos químicos são aplicados nas culturas, visando controlar o surgimento e proliferação de pragas. Assim, faz-se o uso dos praguicidas, dentre eles, o diuron (N-(3,4-diclorofenil)-N,N-dimetiluréia), que é aplicado em plantações ao redor do mundo. O diuron pode ser biodegradado em outros três principais compostos o 3,4-dicloroanilina (DCA), 3,4-diclorofenilureia (DCPMU) e 2,4-diclorofenil-N-metilureia (DCPMU). Normalmente, o diuron é aplicado nas plantações juntamente com os alquilfenóis etoxilatos (APE), como o nonilfenol etoxilato (NPE) e o octilfenol etoxilato (OPE), compostos que facilitam a dispersão do diuron. Alguns estudos têm demonstrado que tanto o diuron quanto os alquilfenóis podem causar alterações enzimáticas e/ou estrogênicas em diversos organismos. Dentre as enzimas que podem ser alteradas, estão as enzimas de biotransformação, tanto de fase I (7-etoxirresorufina-O-deetilase – EROD, 7-pentóxiresorufinaO-desalquilase – PROD, 7-benzilóxiresorufina-O-desalquilase – BROD e a P450 aromatase), quanto as de fase II (glutationa-S-transferase – GST) e as proteínas de fase III (resistência a multixenobioticos – MXR). Outros parâmetros que podem sofrer alterações são os antioxidantes (superóxido dismutase – SOD, catalase – CAT, glutationa peroxidase – GPx, glutationa redutase – GR, glutationa-6-fosfato desidrogenase – G6PDH, aldeído desidrogenase – ALDH e peroxidação lipídica), que são os responsáveis pelo controle entre a produção e o combate às espécies reativas de oxigênio (ERO). Dentre os parâmetros endócrinos que podem ser alterados, e então utilizados como biomarcadores, estão as enzimas CYP3A e a 17β-hidroxiesteróide desidrogenase (17β-HSD) e ainda a proteína vitelogenina (Vtg). Assim o objetivo deste trabalho foi avaliar, em tilápias do Nilo (Oreochromis nilotucus) (1) quais os efeitos do diuron e seus metabólitos, em associação ou não, em diferentes concentrações e (2) quais os efeitos do diuron, seus metabólitos e os alquilfenois, em associação ou não em diferentes concentrações no fígado e nas brânquias de peixes expostos por 7 dias, utilizando os seguintes parametros, EROD, BROD, PROD, GST, MXR, SOD, CAT, GPx, G6PDH e peroxidação lipídica (MDA). Em uma terceira etapa, o objetivo do trabalho foi (3) avaliar parâmetros endócrinos em tilápias mozambique (Oreochromis mossambicus) expostas ao diuron, seus metabólitos e alquilfenóis, em associação ou não, em diferentes concentrações, expostas por 7 dias, em fígado e cérebro dos peixes expostos, analisando os seguintes parâmetros: 17β-HSD, P450 aromatase, CYP3A, vitelogenina e a biotransformação do xenobiótico (in vitro). Nossos resultados demonstraram que, em tilápias do Nilo expostas a esses contaminantes nas concentrações e no tempo de exposição utilizados, todos os biomarcadores analisados sofreram alguma alteração após a exposição, alguns mais expressivos como a EROD, MDA, GPx e G6PDH e outros menos. Os resultados obtidos após a exposição de tilápias mozambique aos contaminantes também nos mostraram alterações, confirmando que esses compostos podem ser considerados desreguladores endócrinos, já que todos os parâmetros endócrinos analisados foram alterados. Assim, os resultados obtidos nos mostraram que realmente o diuron, seus metabólitos e os alquilfenóis podem causar alterações bioquímicas em tilápias do Nilo e mozambique, expostos à concentrações ambientalmente relevantes por sete dias, sendo os metabólitos e os APs os principais compostos a causar alterações nos parâmetros analisados. / Brazil is the main sugar cane producer in the world, and to support these productions many chemical compounds have being apply in agriculture, aiming to control the appearance and proliferation of pests. Therefore, the use of pesticide, as diuron, (N- (3,4-dichlorophenyl) -N,N-dimethylurea), in some crops in the world is common. Diuron can be biodegraded in three others compounds, 3,4-dichloroaniline (DCA), 3,4-dichlorophenylurea (DCPU) and 2,4-dichlorophenyl-N-methylurea (DCPMU). Normally, diuron have being applied associated with alkylphenols ethoxylates (APE), like nonilphenol (NP) and octylphenol (OP), which increase the solubility and dispersion of the herbicide. Some studies have shown that diuron and alkylphenols can cause enzymatic and/or estrogenic changes, in various organisms. Therefore, biotransformation enzymes, phase I (7-ethoxirresorufin-O-deethylase – EROD, 7- pentaxirisorufin-O-dealkylase – PROD, 7-benzyloxysorufin-O-desalkylase – BROD and P450 aromatase), phase II (glutathione-S-transferase-GST) and phase III proteins (multixenobiotic resistance – MXR), and antioxidant parameters (superoxide dismutase – SOD, catalase – CAT, glutathione peroxidase – GPx, glutathione reductase – GR, glutathione-6-phosphate dehydrogenase – G6PDH, aldehyde dehydrogenase – ALDH and lipid peroxidation), which are responsible for the control between production and degradation of reactive oxygen species (ROS). The endocrine parameters that can be altered and used as biomarkers are the CYP3A and 17β-hydroxysteroid dehydrogenase (17β-HSD) enzymes and the vitellogenin (Vtg) protein. All of these parameters can be altered after exposure of organisms to xenobiotics. Therefore, the aim of this work was evaluate the effects of (1) diuron and its metabolites, alone or in combination, in different concentrations, and the effect of (2) diuron, its metabolites and the alkyphenols, alone or in combination, in different concentration, in liver and gill of Nile tilapia (Oreochromis niloticus) exposed for 7 days, using as parameters EROD, BROP, PROD, GST, MXR, SOD, CAT, GPx, G6PDH and lipid peroxidation (MDA). In a third step, the aim of the work as (3) evaluate endocrine parameters in mossambique tilapia (Oreochromis mossambicus), exposed to diuron, its metabolites and alkyphenols, alone or in combination, in different concentrations for 7 days, in liver and brain, using as parameters 17β-HSD, CYP3A, vitellogenin and xenobiotic biotransformation (in vitro). Our results demonstrated that all contaminants, at the concentration and time of exposure, can cause alteration in all biomarkers, some more than others, as EROD, MDA, GPx and G6PDH, in Nile tilapia, and the results obtained after mozambique tilapia exposure, to the contaminants, also showed changes, confirming that these compounds can be considered endocrine disruptors. Thus, the results showed that diuron, its metabolites and alkylphenols could cause biochemical changes in fish, Nile and mossambique tilapia, exposing to the environmentally relevant concentrations for seven days, being diruon metabolites and APs the main compounds to cause changes in the analyzed parameters. / FAPESP: 2014/18825-9

Cette thèse est principalement consacrée à l'association des méthodes de points intérieurs et des techniques de l'optimisation DC et DCA pour résoudre les problèmes d'optimisation non convexes de grande taille.<br />La thèse comporte trois parties : <br />la première partie est consacrée aux techniques d'optimisations locales et s'articule autour des méthodes de points intérieurs et de la programmation DC. Nous y développons deux algorithmes. Après une présentation non exhaustive de la programmation DC, des méthodes de points intérieurs et des propriétés essentielles de la classe des matrices quasi-définies au chapitre un, nous présentons au chapitre deux un nouvel algorithme basé sur une reformulation des conditions d'optimalité de Karush-Kuhn-Tucker. Le troisième chapitre est consacré à l'intégration des techniques d'optimisation DC dans un schéma de points intérieurs, c'est l'algorithme IPDCA.<br />La seconde partie de la thèse est consacrée aux solutions globales de problèmes de programmation quadratique. Dans le premier chapitre de cette partie nous explorons l'intégration d'IPDCA dans un schéma B&B. Le second chapitre de la partie est consacré à la résolution de problèmes quadratiques à variables 0-1 par un schéma B\&B dans lequel nous faisons intervenir IPDCA. Le troisième chapitre est quant à lui consacré à l'optimisation monotone due au Professeur Tuy. Nous examinons plus particulièrement son intégration dans un B&B dans lequelle DCA est appelé pour améliorer la borne supérieure.<br />Le quatrième et dernier chapitre de cette partie est consacré à une procédure de redémarrage de DCA. <br />La dernière partie de la thèse est consacrée aux applications industrielles. Nous y appliquons les deux algorithmes développés dans la première partie de la thèse à un problème de mécanique de structure de grande dimension et à un problème en Data Mining.

No description available.

Production of oil and gas from shale is often described as a revolution to energyproduction in North America. Since the beginning of this century the shale oilproduction has increased from practically zero to currently supply almost half of theU.S. oil production. This development is made possible by the technology ofhorizontal drilling and hydraulic fracturing. Since the production has not been ongoingfor that long, production data is still fairly limited in length and there are still largeuncertainties in many parameters, for instance production decline, lifespan, drainagearea, geographical extent and future technological development. More research isneeded to be able to estimate future production and resources with more certainty. At the moment shale oil is extracted only in North America but around the worldinvestigations are starting to assess if the conditions are suitable from shale oilextraction elsewhere. The global technically recoverable resource has been estimatedto 345 Gb, 10% of all global technically recoverable resources. Health andenvironmental aspects of shale oil and gas production have not yet been investigatedthoroughly and there is a risk that these parameters may slow down or limit thespreading of shale development. This report aims to examine production patterns of shale oil wells by applying declinecurve analysis. This analysis comprises of analyzing historical production data toinvestigate how the future production may develop. The area of the study is the EagleFord shale play in Texas, U.S. The goal is to fit decline curves to production data andthen use them for making estimates of future production in the Eagle Ford. The production in the shale oil wells included in the study reach their peak already within a few months after production starts. After this point, production is declining.After one year, production has decreased by 75% and after two years the productionis 87% of the peak production. The hyperbolic decline curve has a good fit toproduction data and in many cases the curve is close to harmonic. It is too early todetermine whether the alternative decline curve that is tested, the scaling declinecurve, has a better fit in the long term. The report also investigates how the density of the petroleum affects the declinecurve. The result is that lighter products decline faster than heavier. A sensitivity analysis is performed to illustrate how different parameters affect thefuture production development. In addition to the wells’ decline rate, the assumptionson the maximum number of wells, the maximal production and the rate at which newwells are added affect the ultimately recoverable resource. These parameters all havelarge uncertainties and makes resource estimations more difficult.

Cette thèse s'articule autour des fonctions booléennes liées à la cryptographie et la cryptanalyse de certains schémas d'identification. Les fonctions booléennes possèdent des propriétés algébriques fréquemment utilisées en cryptographie pour constituer des S-Boxes (tables de substitution).Nous nous intéressons, en particulier, à la construction de deux types de fonctions : les fonctions courbes et les fonctions équilibrées de haut degré de non-linéarité.Concernant la cryptanalyse, nous nous focalisons sur les techniques d'identification basées sur les problèmes de perceptron et de perceptron permuté. Nous réalisons une nouvelle attaque sur le schéma afin de décider de sa faisabilité.Nous développons ici des nouvelles méthodes combinant l'approche déterministe DCA (Difference of Convex functions Algorithm) et heuristique (recuit simulé, entropie croisée, algorithmes génétiques...). Cette approche hybride, utilisée dans toute cette thèse, est motivée par les résultats intéressants de la programmation DC.

The robust glycolytic metabolism of glioblastoma multiforme (GBM) has proven them susceptible to increases in oxidative metabolism induced by the pyruvate mimetic dichloroacetate (DCA). Recent reports demonstrate that the anti-diabetic drug metformin enhances the damaging oxidative stress associated with DCA treatment in cancer cells. We sought to elucidate the role of metformin’s reported activity as a mitochondrial complex I inhibitor in the enhancement of DCA cytotoxicity in the VM-M3 model of GBM. We demonstrated that metformin potentiated DCA-induced superoxide production and that this was required for enhanced cytotoxicity towards VM-M3 cells with the combination. Similarly, rotenone enhanced oxidative stress resultant from DCA treatment and this too was required for the noted augmentation of cytotoxicity. Adenosine monophosphate kinase (AMPK) activation was not observed with the concentration of metformin required to enhance DCA activity. Moreover, addition of an activator of AMPK did not enhance DCA cytotoxicity, whereas an inhibitor of AMPK heightened the cytotoxicity of the combination. We also show that DCA and metformin reduce tumor burden and prolong survival in VM-M3 tumor-burdened mice as individual therapies. In contrast to our in vitro work, we did not observe synergy between DCA and metformin in vivo. Our data indicate that metformin enhancement of DCA cytotoxicity is dependent on complex I inhibition. Particularly, that complex I inhibition cooperates with DCA-induction of glucose oxidation to enhance cytotoxic oxidative stress in VM-M3 GBM cells. This work supports further investigation and optimization of a DCA/metformin combination as a potential pro-oxidant combinatorial therapy for GBM.