Tesis sobre el tema "Caractéristiques aléatoires"
Crea una cita precisa en los estilos APA, MLA, Chicago, Harvard y otros
Consulte los 23 mejores tesis para su investigación sobre el tema "Caractéristiques aléatoires".
Junto a cada fuente en la lista de referencias hay un botón "Agregar a la bibliografía". Pulsa este botón, y generaremos automáticamente la referencia bibliográfica para la obra elegida en el estilo de cita que necesites: APA, MLA, Harvard, Vancouver, Chicago, etc.
También puede descargar el texto completo de la publicación académica en formato pdf y leer en línea su resumen siempre que esté disponible en los metadatos.
Explore tesis sobre una amplia variedad de disciplinas y organice su bibliografía correctamente.
Le, Khanh Toan. "Identification des caractéristiques aléatoires de remblais à partir du suivi de santé des structures : application aux structures portuaires". Nantes, 2012. http://www.theses.fr/2012NANT2021.
Texto completoThe in-service behavior of some prototype structures (like Civil Engineering Structures) is often different from that initially assume during the design. Even if assumptions during design are conservative they hold the understanding of the real behavior off. The modeling of in-service behavior is of first importance when reassessing complex structures like harbor structures and performing risk analysis. To this aim, the Structural Health Monitoring (SHM) is of first importance because it allows us to assess the real level of loading and to provide more realistic models for mechanical behavior or value of their parameters. This thesis proposes to use information from instrumentation in a probabilistic context. This is to identify parameters of a behavior model under form of random variables represented by a vector X which integrate in part the natural hazard and another for a model error. The research is performed in two stages: (1) simulation of the structure behavior by using a set of a priori parameters as input value of the finite element code (PLAXIS) on a simple model M (X) which is calibrated by a quadratic polynomial response surface model RS (X) for points well chosen. This approach allow to converge to the same asymptotic properties for realizations of X not sampled in the base of calibration. This reduces the sample size, which represents a better advantage than using Bayesian or Neuronal networks; (2) an inverse analysis is realized on a large number of situations measured to identify X and to give Y that Y= RS (X). We propose to apply this method to the instrumentation of pile supported wharfs to determine site characteristics, previously unavailable in probabilistic models
Bernard, Simon. "Forêts Aléatoires: De l'Analyse des Mécanismes de Fonctionnement à la Construction Dynamique". Phd thesis, Université de Rouen, 2009. http://tel.archives-ouvertes.fr/tel-00598441.
Texto completoBernard, Simon. "Forêts aléatoires : de l’analyse des mécanismes de fonctionnement à la construction dynamique". Phd thesis, Rouen, 2009. http://www.theses.fr/2009ROUES011.
Texto completoThis research work is related to machine learning and more particularlydealswiththeparametrizationofRandomForests,whichareclassifierensemble methods that use decision trees as base classifiers. We focus on two important parameters of the forest induction : the number of features randomly selected at each node and the number of trees. We first show that the number of random features has to be chosen regarding to the feature space properties, and we propose hence a new algorithm called Forest-RK that exploits those properties. We then show that a static induction process implies that some of the trees of the forest make the ensemble generalisation error decrease, by deteriorating the strength/correlation compromise. We finaly propose an original random forest dynamic induction algorithm that favorably compares to static induction processes
Desbordes, Paul. "Méthode de sélection de caractéristiques pronostiques et prédictives basée sur les forêts aléatoires pour le suivi thérapeutique des lésions tumorales par imagerie fonctionnelle TEP". Thesis, Normandie, 2017. http://www.theses.fr/2017NORMR030/document.
Texto completoRadiomics proposes to combine image features with those extracted from other modalities (clinical, genomic, proteomic) to set up a personalized medicine in the management of cancer. From an initial exam, the objective is to anticipate the survival rate of the patient or the treatment response probability. In medicine, classical statistical methods are generally used, such as theMann-Whitney analysis for predictive studies and analysis of Kaplan-Meier survival curves for prognostic studies. Thus, the increasing number of studied features limits the use of these statistics. We have focused our works on machine learning algorithms and features selection methods. These methods are resistant to large dimensions as well as non-linear relations between features. We proposed two features selection strategy based on random forests. Our methods allowed the selection of subsets of predictive and prognostic features on 2 databases (oesophagus and lung cancers). Our algorithms showed the best classification performances compared to classical statistical methods and other features selection strategies studied
Daher, Ziad. "Estimation de grandeurs caractéristiques des vibrations des machines tournantes en régime variable". Thesis, Orléans, 2011. http://www.theses.fr/2011ORLE2059.
Texto completoVibration analysis is one of the main methods for rotating machinery monitoring. The topic of this thesis is the diagnosis of rolling element bearings under varying rotating speed. The objective is the development of software sensors providing a relevant indicator of bearing condition monitoring from the vibration signal generated by the bearing. Often, rotating machinery is composed from gears and bearings. Gear faults generate deterministic components and bearing faults generate random components. To diagnose, the key point is to extract the vibration signal caused by bearing from the total vibration signal. In constant rotating speed condition, the used tool to estimate the deterministic component is the time synchronous averaging. In varying rotating speed, the signal becomes "cyclo-non-stationary". Therefore, the traditional synchronous averaging method becomes inappropriate and one needs to extend its definition and estimation process. In this thesis, we develop a new approach based on a decomposition of the vibration signal into a set of cyclo-non-stationary components. This allowed us to develop a tool for estimating the deterministic components of the signal which are characterized by cyclo-non-stationarity order 1 and to develop a tool for characterizing the cyclo-non-stationarity order 2 based on synchronized spectral correlation. The performance of this tool has been tested on synthetic signals and on real signals
Vathanakhool, Khoollapath. "Estimation de la sécurité des poteaux en béton armé : compte tenu des variations aléatoires de leurs caractéristiques géométriques et mécaniques le long de leur ligne moyenne". Toulouse, INSA, 1987. http://www.theses.fr/1987ISAT0015.
Texto completoSokolovska, Nataliya. "Contributions à l'estimation de modèles probabilistes discriminants : apprentissage semi-supervisé et sélection de caractéristiques". Phd thesis, Ecole nationale supérieure des telecommunications - ENST, 2010. http://tel.archives-ouvertes.fr/tel-00557662.
Texto completoSokolovska, Nataliya. "Contributions à l'estimation de modèles probabilistes discriminants : apprentissage semi-supervisé et sélection de caractéristiques". Phd thesis, Paris, Télécom ParisTech, 2010. https://pastel.hal.science/pastel-00006257.
Texto completoIn this thesis, we investigate the use of parametric probabilistic models for classification tasks in the domain of natural lang uage processing. We focus in particular on discriminative models, such as logistic regression and its generalization, conditional random fields (CRFs). Discriminative probabilistic models design directly conditional probability of a class given an observation. The logistic regression has been widely used due to its simplicity and effectiveness. Conditional random fields allow to take structural dependencies into consideration and therefore are used for structured output prediction. In this study, we address two aspects of modern machine learning, namely , semi-supervised learning and model selection, in the context of CRFs. The contribution of this thesis is twofold. First, we consider the framework of semi -supervised learning and propose a novel semi-supervised estimator and show that it is preferable to the standard logistic regression. Second, we study model selection approaches for discriminative models, in particular for CRFs and propose to penalize the CRFs with the elastic net. Since the penalty term is not differentiable in zero, we consider coordinate-wise optimization. The comparison with the performances of other methods demonstrates competitiveness of the CRFs penalized by the elastic net
Wacker, Jonas. "Random features for dot product kernels and beyond". Electronic Thesis or Diss., Sorbonne université, 2022. http://www.theses.fr/2022SORUS241.
Texto completoDot product kernels, such as polynomial and exponential (softmax) kernels, are among the most widely used kernels in machine learning, as they enable modeling the interactions between input features, which is crucial in applications like computer vision, natural language processing, and recommender systems. However, a fundamental drawback of kernel-based statistical models is their limited scalability to a large number of inputs, which requires resorting to approximations. In this thesis, we study techniques to linearize kernel-based methods by means of random feature approximations and we focus on the approximation of polynomial kernels and more general dot product kernels to make these kernels more useful in large scale learning. In particular, we focus on a variance analysis as a main tool to study and improve the statistical efficiency of such sketches
Marchina, Bastien. "Tests d'adéquation basés sur la fonction caractéristique". Thesis, Montpellier 2, 2011. http://www.theses.fr/2011MON20088/document.
Texto completoThis PhD thesis consists in building goodness-of-fit tests using the characteristic function (CF) as a prefered representation for the probability laws involved.We start with listing and improving results in probability theory necessary to build test statistics using the characteristic function and its conterpart the empirical characteristic function.We list and classify existing characteristic function based goodness-of-fit tests published by varions authors since 1977.Then, we build a class of tests based on integral metrics. We take particular attention to the case where the statistics are build using a L2 distance. More specifically, we give asymptotic results in this case. However, these results reveal the need for information on the unknown eigenelements of an integral operator. Thus, we present and implement an approximation method using a sequence of projections on orthonormal bases ofan hilbertian functional space.Finally, we will build another class of tests using the Neyman smooth test paradigm. This study is based on our previous results, that fit well into the construction of characteristic function based smooth tests. We give various applications, presenting tests for both a simple hypothesis and a composite hypothesis
Ferraz, Eduardo. "Topologie algébrique de complexes simpliciaux aléatoires et applications aux réseaux de capteurs". Thesis, Paris, ENST, 2012. http://www.theses.fr/2012ENST0006/document.
Texto completoThis thesis has two main parts. Part I uses stochastic anlysis to provide bounds for the overload probability of different systems thanks to concentration inequalities. Although the results are general, we apply them to real wireless network systems such as WiMax and mutliclass user traffic in an OFDMA system. In part I I, we find more connections between the topology of the coverage of a sensor network and the topology of its corresponding simplicial complex. These connections highlight new aspects of Betti numbers, the number of k-simplices, and Euler characteristic. Then, we use algebraic topology in conjunction with stochastic analysis, after assuming that the positions of the sensors are points of a Point point process. As a consequence we obtain, in d dimensions, the statistics of the number of k-simplices and of Euler characteristic, as well as upper bounds for the distribution of Betti numbers. We also prove that the number of k-simplices tends to a Gaussian distribution as the density of sensors grows, and we specify the convergence rate. Finally, we restrict ourselves to one dimension. In this case, the problem becomes equivalent to solving a M/M/1/1 preemptive queue. We obtain analytical results for quantites such as the distribution of the number of connected components and the probability of complete coverage
Letendre, Thomas. "Contributions à l'étude des sous-variétés aléatoires". Thesis, Lyon, 2016. http://www.theses.fr/2016LYSE1240/document.
Texto completoWe study the volume and Euler characteristic of codimension r ∈ {1, . . . , n} random submanifolds in a dimension n manifold M. First, we consider Riemannian random waves. That is M is a closed Riemannian manifold and we study the common zero set Zλ of r independent random linear combinations of eigenfunctions of the Laplacian associated to eigenvalues smaller than λ 0. We compute estimates for the mean volume and Euler characteristic of Zλ as λ goes to infinity. We also consider a model of random real algebraic manifolds. In this setting, M is the real locus of a projective manifold defined over the reals. Then, we consider the real vanishing locus Zd of a random real global holomorphic section of E ⊗ Ld, where E is a rank r Hermitian vector bundle, L is an ample Hermitian line bundle and both these bundles are defined over the reals. We compute the asymptotics of the mean volume and Euler characteristic of Zd as d goes to infinity. In this real algebraic setting, we also compute the asymptotic of the variance of the volume of Zd, when 1 r < n. In this case, we prove asympotic equidistribution results for Zd in M
Ferraz, Eduardo. "Topologie algébrique de complexes simpliciaux aléatoires et applications aux réseaux de capteurs". Electronic Thesis or Diss., Paris, ENST, 2012. http://www.theses.fr/2012ENST0006.
Texto completoThis thesis has two main parts. Part I uses stochastic anlysis to provide bounds for the overload probability of different systems thanks to concentration inequalities. Although the results are general, we apply them to real wireless network systems such as WiMax and mutliclass user traffic in an OFDMA system. In part I I, we find more connections between the topology of the coverage of a sensor network and the topology of its corresponding simplicial complex. These connections highlight new aspects of Betti numbers, the number of k-simplices, and Euler characteristic. Then, we use algebraic topology in conjunction with stochastic analysis, after assuming that the positions of the sensors are points of a Point point process. As a consequence we obtain, in d dimensions, the statistics of the number of k-simplices and of Euler characteristic, as well as upper bounds for the distribution of Betti numbers. We also prove that the number of k-simplices tends to a Gaussian distribution as the density of sensors grows, and we specify the convergence rate. Finally, we restrict ourselves to one dimension. In this case, the problem becomes equivalent to solving a M/M/1/1 preemptive queue. We obtain analytical results for quantites such as the distribution of the number of connected components and the probability of complete coverage
Kozyrskiy, Bogdan. "Exploring the Intersection of Bayesian Deep Learning and Gaussian Processes". Electronic Thesis or Diss., Sorbonne université, 2023. https://accesdistant.sorbonne-universite.fr/login?url=https://theses-intra.sorbonne-universite.fr/2023SORUS064archi.pdf.
Texto completoDeep learning played a significant role in establishing machine learning as a must-have instrument in multiple areas. The use of deep learning poses several challenges. Deep learning requires a lot of computational power for training and applying models. Another problem with deep learning is its inability to estimate the uncertainty of the predictions, which creates obstacles in risk-sensitive applications. This thesis presents four projects to address these problems: We propose an approach making use of Optical Processing Units to reduce energy consumption and speed up the inference of deep models. We address the problem of uncertainty estimates for classification with Bayesian inference. We introduce techniques for deep models that decreases the cost of Bayesian inference. We developed a novel framework to accelerate Gaussian Process regression. We propose a technique to impose meaningful functional priors for deep models through Gaussian Processes
Hardy, Adrien. "Problèmes d'équilibre vectoriels et grandes déviations". Toulouse 3, 2013. http://thesesups.ups-tlse.fr/2210/.
Texto completoIn this thesis we investigate the convergence and large deviations of the empirical measure associated with several determinantal point processes. These point processes have in common that their average characteristic polynomial is a multiple orthogonal polynomial, the latter being a generalization of orthogonal polynomials. The first simplest example is a 2D Coulomb gas in a confining potential at inverse temperature beta = 2, for which the average characteristic polynomial is an orthogonal polynomial. A large deviation principle for the empirical measure is known to hold, even in the general beta > 0 case, with a rate function involving an equilibrium problem arising from logarithmic potential theory. As a warming up, we show this result actually extends to the case where the potential is weakly confining, i. E. Satisfying a weaker growth assumption that usual. To do so, we introduce a compactification procedure which will be of important use in what follows. Motivated by more complex determinantal point processes, we then develop a general framework for vector equilibrium problems with weakly confining potentials to make sense. We prove existence and uniqueness of their solutions, which improves the existing results in the potential theory literature, and moreover show that the associated functionals have compact level sets. Next, we investigate a determinantal point process associated with an additive perturbation of a Wishart matrix, for which the average characteristic polynomial is a multiple orthogonal polynomial associated with two weights. We establish a large deviation principle for the empirical measure with a rate function related to a vector equilibrium problem with weakly confining potentials. This is the first time that a vector equilibrium problem is shown to be involved in a large deviation principle for random matrix models. Finally, we study on a more general level when both the empirical measure of a determinantal point process and the zero distribution of the associated average characteristic polynomial converge to the same limit. We obtain a sufficient condition for a class of determinantal point processes which contains the ones related to multiple orthogonal polynomials. On the way, we provide a sufficient condition to strengthen the mean convergence of the empirical measure to the almost sure one. As an application, we describe the limiting distributions for the zeros of multiple Hermite and multiple Laguerre polynomials in terms of free convolutions of classical distributions with atomic measures, and then derive algebraic equations for their Cauchy-Stieltjes transforms
Desir, Chesner. "Classification Automatique d'Images, Application à l'Imagerie du Poumon Profond". Phd thesis, Université de Rouen, 2013. http://tel.archives-ouvertes.fr/tel-00879356.
Texto completoDesir, Chesner. "Classification automatique d'images, application à l'imagerie du poumon profond". Phd thesis, Rouen, 2013. http://www.theses.fr/2013ROUES053.
Texto completoThis thesis deals with automated image classification, applied to images acquired with alveoscopy, a new imaging technique of the distal lung. The aim is to propose and develop a computer aided-diagnosis system, so as to help the clinician analyze these images never seen before. Our contributions lie in the development of effective, robust and generic methods to classify images of healthy and pathological patients. Our first classification system is based on a rich and local characterization of the images, an ensemble of random trees approach for classification and a rejection mechanism, providing the medical expert with tools to enhance the reliability of the system. Due to the complexity of alveoscopy images and to the lack of expertize on the pathological cases (unlike healthy cases), we adopt the one-class learning paradigm which allows to learn a classifier from healthy data only. We propose a one-class approach taking advantage of combining and randomization mechanisms of ensemble methods to respond to common issues such as the curse of dimensionality. Our method is shown to be effective, robust to the dimension, competitive and even better than state-of-the-art methods on various public datasets. It has proved to be particularly relevant to our medical problem
El-Haddad, Mireille. "Schéma de transport de l'interface d'un écoulement diphasique visqueux non miscible par la méthode des caractéristiques". Thesis, Paris 6, 2016. http://www.theses.fr/2016PA066475/document.
Texto completoIn this thesis, we use mathematical and numerical tools to model three dimensional incompressible laminar flows with free surface. The described industrial application is the study of the mould filling phase in a foundry. We develop an algorithm for the transport of the interface by the fluid velocity for a viscous incompressible immiscible fluids of high density ratio in two-phase flow using the method of characteristics for the convection problem.There are, however, major challenges in the context of two-phase flow modeling.First, we have to take into account the evolution of the interface and its topological changes. Second, we have to deal with the non-linearity for the convection of the flow and the interface. Third, we must assign appropriate boundary conditions to the flow and transport equations.In addition, care must be taken in treating the geometrical and topological singularities across the interface.We also have to maintain a sharp interface resolution, including the cases of interface folding, breaking and merging.Furthermore, we should respect the physical properties such as the mass conservation for any incompressible fluid flows.Finally, we have to keep in mind the limitations in the time of computation and memory to solve this kind of problem in practical cases. Our purpose is to find a reliable scheme able to model the filling of three dimensional industrial moulds.The first part of the thesis is devoted to the mathematical description of the interface transport scheme by the fluid velocity. The fluids motion is described by the Navier-Stokes equations. The interface is captured by the Level-Set function. The problem isdiscretized by the characteristics method in time and finiteelements method in space. The interface is captured by the Level-Setfunction. Appropriate boundary conditions for the problem ofmould filling are investigated, a new natural boundary conditionunder pressure effect for the transport equation is proposed andan algorithm for computing the solution is presented. Finally,numerical experiments show and validate the effectiveness of theproposed scheme.In the second part of the thesis, we introduce a domain decomposition method that suits the discretization by the method of characteristics in order to improve the performance of the proposed algorithm to model the filling phase for moulds of medium series. Numerical results of comparison validate the precision of the parallel code
Lafaye, de Micheaux Pierre. "Tests d'indépendance en analyse multivariée et tests de normalité dans les modèles ARMA". Phd thesis, Université Montpellier II - Sciences et Techniques du Languedoc, 2002. http://tel.archives-ouvertes.fr/tel-00002192.
Texto completopour les innovations d'un modèle ARMA(p,q) de tendance et moyenne
connues, basé sur l'approche du test lisse dépendant des
données et simple à appliquer. Une vaste étude de simulation
est menée pour étudier ce test pour des
tailles échantillonnales modérées. Notre approche
est en général plus puissante que les tests existants. Le niveau est
tenu sur la majeure partie de l'espace paramétrique. Cela est en accord avec les résultats
théoriques montrant la supériorité de l'approche du test lisse
dépendant des données dans des contextes similaires.
Un test d'indépendance (ou d'indépendance sérielle) semi-paramétrique entre des sous-vecteurs de loi normale est proposé, mais sans
supposer la normalité jointe de ces marginales. La statistique de test
est une fonctionnelle de type Cramér-von Mises d'un processus défini à
partir de la fonction caractéristique empirique. Ce processus est
défini de façon similaire à celui de Ghoudi et al. (2001) construit à
partir de la fonction de répartition empirique et utilisé pour tester
l'indépendance entre des marginales univariées. La statistique de test
peut être représentée comme une V-statistique. Il est convergent pour
détecter toute forme de dépendance. La convergence faible du processus
est établie. La distribution asymptotique des fonctionnelles de
Cramér-von Mises est approchée par la méthode de Cornish-Fisher au
moyen d'une formule de récurrence pour les cumulants et par le
calcul numérique des valeurs propres dans la formule d'inversion. La
statistique de test est comparée avec celle de Wilks pour
l'hypothèse paramétrique d'indépendance dans le modèle MANOVA à
un facteur avec effets aléatoires.
Augustyniak, Maciej. "Une famille de distributions symétriques et leptocurtiques représentée par la différence de deux variables aléatoires gamma". Thèse, 2008. http://hdl.handle.net/1866/8192.
Texto completoPanneton, François. "Construction d'ensembles de points basée sur des récurrences linéaires dans un corps fini de caractéristique 2 pour la simulation Monte Carlo et l'intégration quasi-Monte Carlo". Thèse, 2004. http://hdl.handle.net/1866/16634.
Texto completoPrévost, Noémie. "Inférence topologique". Thèse, 2014. http://hdl.handle.net/1866/10745.
Texto completoData coming from a fine sampling of a continuous process (random field) can be represented as images. A statistical test aiming at detecting a difference between two images can be seen as a group of tests in which each pixel is compared to the corresponding pixel in the other image. We then use a method to control the type I error over all the tests, such as the Bonferroni correction or the control of the false discovery rate (FDR). Methods of data analysis have been developped in the field of medical imaging, mainly by Keith Worsley, using the geometry of random fields in order to build a global statistical test over the whole image. The expected Euler characteristic of the excursion set of the random field underlying the sample over a given threshold is used in order to determine the probability that the random field exceeds this same threshold under the null hypothesis (topological inference). We present some notions relevant to random fields, in particular isotropy (the covariance function between two given points of a field depends only on the distance between them). We discuss two methods for the analysis of non\-isotropic random fields. The first one consists in deforming the field and then using the intrinsic volumes and the Euler characteristic densities. The second one uses the Lipschitz-Killing curvatures. We then perform a study of sensitivity and power of the topological inference technique comparing it to the Bonferonni correction. Finally, we use topological inference in order to describe the evolution of climate change over Quebec territory between 1991 and 2100 using temperature data simulated and published by the Climate Simulation Team at Ouranos, with the Canadian Regional Climate Model CRCM4.2.
Kadje, Kenmogne Romain. "Estimation de paramètres en exploitant les aspects calculatoires et numériques". Thèse, 2017. http://hdl.handle.net/1866/20584.
Texto completo