Les actions de groupes en géométrie symplectique et l'application moment

Ce mémoire porte sur quelques notions appropriées d'actions de groupe sur les variétés symplectiques, à savoir en ordre décroissant de généralité : les actions symplectiques, les actions faiblement hamiltoniennes et les actions hamiltoniennes. Une connaissance des actions de groupes et de la géométrie symplectique étant prérequise, deux chapitres sont consacrés à des présentations élémentaires de ces sujets. Le cas des actions hamiltoniennes est étudié en détail au quatrième chapitre : l'importante application moment y est définie et plusieurs résultats concernant les orbites de la représentation coadjointe, tels que les théorèmes de Kirillov et de Kostant-Souriau, y sont démontrés. Le dernier chapitre se concentre sur les actions hamiltoniennes des tores, l'objectif étant de démontrer le théorème de convexité d'Atiyha-Guillemin-Sternberg. Une discussion d'un théorème de classification de Delzant-Laudenbach est aussi donnée. La présentation se voulant une introduction assez exhaustive à la théorie des actions hamiltoniennes, presque tous les résultats énoncés sont accompagnés de preuves complètes. Divers exemples sont étudiés afin d'aider à bien comprendre les aspects plus subtils qui sont considérés. Plusieurs sujets connexes sont abordés, dont la préquantification géométrique et la réduction de Marsden-Weinstein.

A Coupled 2 × 2D Babcock–Leighton Solar Dynamo Model. I. Surface Magnetic Flux Evolution

The need for reliable predictions of the solar activity cycle motivates the development of dynamo models incorporating a representation of surface processes sufficiently detailed to allow assimilation of magnetographic data. In this series of papers we present one such dynamo model, and document its behavior and properties. This first paper focuses on one of the model's key components, namely surface magnetic flux evolution. Using a genetic algorithm, we obtain best-fit parameters of the transport model by least-squares minimization of the differences between the associated synthetic synoptic magnetogram and real magnetographic data for activity cycle 21. Our fitting procedure also returns Monte Carlo-like error estimates. We show that the range of acceptable surface meridional flow profiles is in good agreement with Doppler measurements, even though the latter are not used in the fitting process. Using a synthetic database of bipolar magnetic region (BMR) emergences reproducing the statistical properties of observed emergences, we also ascertain the sensitivity of global cycle properties, such as the strength of the dipole moment and timing of polarity reversal, to distinct realizations of BMR emergence, and on this basis argue that this stochasticity represents a primary source of uncertainty for predicting solar cycle characteristics.

Étude descriptive des perceptions de soutien, d’habilitation et de reconnaissance favorisant la mobilisation organisationnelle au Ministère de l’éducation nationale et de l’alphabétisation du Burkina Faso

Cette thèse de doctorat se situe dans le contexte des préoccupations des autorités nationales et des organisations internationales relatives à l’efficacité des organisations publiques dans les pays en développement, dans le cadre des Objectifs du Millénaire pour le développement (OMD) et du programme Éducation pour tous (ÉPT). L’argument du manque de ressources est de plus en plus remis en cause par le constat que certaines organisations disposant de ressources moindres que d’autres de même nature, obtiennent de meilleurs résultats (Barney, 1991; Durand, 1996; Isckia, 2008). Autrement dit, la quantité de ressources n’explique pas tout; il faut considérer d’autres éléments, dont la mobilisation organisationnelle, c’est-à-dire le mouvement obtenu d’une masse critique d’employés qui adoptent des actions positives dans le sens de l’atteinte des objectifs de leur organisation. Cette mobilisation suppose un climat positif auquel contribue la présence de certains états psychologiques ressentis par les employés, notamment les perceptions de soutien et de reconnaissance de la part de l’organisation de même qu’un sentiment d’habilitation psychologique (Tremblay et Simard, 2005). Ces perceptions et ce sentiment constituent les points focaux de la recherche que nous avons menée au sein du Ministère de l’éducation nationale et de l’alphabétisation (MÉNA) du Burkina Faso. L’objectif principal de notre recherche est de décrire ces trois états psychologiques. Le soutien organisationnel perçu (SOP), l’habilitation psychologique (HP) et la reconnaissance perçue ont été explorés à partir des travaux de Eisenberger et al. (1986), de Spreitzer (1995) et de Brun et Dugas (2005) respectivement. Nous avons délibérément choisi la perspective des employés plutôt que celle des pratiques de gestion observées ou déclarées de leurs supérieurs et avons entrepris de connaître leurs perceptions. Ces dernières méritent que l’on s’en préoccupe car aucune politique, mesure ou pratique visant à instaurer un climat organisationnel mobilisant ne peut être efficace si elle n’est pas perçue comme telle par les employés. Utilisant une méthodologie mixte, nous avons recueilli auprès de cadres et de directions d’école, des données sur les trois états psychologiques retenus, à l’aide d’un questionnaire comportant 37 énoncés (65 répondants); d’entrevues individuelles visant à enrichir, compléter, expliciter ou illustrer les informations obtenues par le questionnaire (18 participants); et de deux groupes de discussion autour des résultats de l’analyse préliminaire des réponses au questionnaire (7 participants). Au total, les données ont été recueillies auprès de 73 personnes, certaines d’entre elles ayant à la fois répondu au questionnaire et participé à une entrevue individuelle. Les données ont été traitées par état psychologique à l’aide des logiciels SPSS Statistics 20 (pour les questionnaires) et QDA Miner 4.0.11 (pour les entretiens individuels). Pour chaque énoncé, chaque variable créée et chacune des caractéristiques (fonction, genre et milieu de travail), nous avons d’abord obtenu des mesures de tendances centrales; nous avons poursuivi en ajoutant un second niveau de traitement en combinant les caractéristiques, par exemple : la fonction (cadre ou direction d’école) et le genre (femme ou homme). Nous avons ensuite procédé au codage des verbatims des entretiens en vue d’en extraire des éléments qui corroborent, précisent ou nuancent les résultats de l’analyse des données obtenues par le questionnaire pour chaque état psychologique. L’exploitation des données d’entretiens visait également à identifier des éléments portant sur le thème de la mobilisation au MÉNA. Les résultats des analyses des données issues des questionnaires indiquent globalement que le soutien organisationnel est perçu de façon négative au sein du MÉNA, seule la valorisation du travail réalisé recueillant un sentiment un peu moins négatif. Les répondants se perçoivent habilités psychologiquement; des quatre composantes de l’habilitation psychologique, c’est l’autonomie qui recueille la perception la plus négative. En ce qui concerne la reconnaissance, on observe une perception positive des éléments reliés à la communication et négative lorsqu’il s’agit de l’appréciation du système d’attribution des récompenses. En complément à ces résultats obtenus de l’analyse des données issues du questionnaire, les entretiens ont permis de mettre en lumière le fait que des pratiques efficaces de gestion (par exemple la rapidité des réponses aux demandes, l’application des normes de ponctualité et d’assiduité, la réception régulière du salaire) sont considérées comme des marques de soutien organisationnel. Ces entretiens ont également permis de découvrir un fort potentiel d’implication personnelle et professionnelle des participants rencontrés qui affichaient une disposition favorable à l’augmentation de leur contribution tout en souhaitant une plus grande reconnaissance de leur potentiel. La prise en compte de la fonction, du genre et du milieu de travail a permis de raffiner les analyses. À titre d’exemple: le soutien organisationnel est perçu plus positivement en milieu urbain qu’en milieu rural; les cadres perçoivent positivement la valorisation que le ministère accorde à leur contribution; alors que les directrices d’école ont une appréciation négative de cette valorisation. Le sentiment de compétence est éprouvé de façon plus positive chez les cadres alors que le sentiment d’autonomie est plus positif chez les directions d’école. En milieu urbain, la transmission de l’information, les rencontres avec les supérieurs et les témoignages d’appréciation sont plus présents qu’en milieu rural. Dans cette recherche, qui s’est déroulée dans un contexte subsaharien, nous avons affiché un parti pris pour une approche universaliste plutôt que culturaliste. Tout en reconnaissant que les traditions et la culture font partie de l’environnement organisationnel, nous pensons qu’elles ne sont pas les principaux facteurs explicatifs des comportements des employés dans une organisation. Les propos tenus par certains des participants que nous avons rencontrés renforcent notre conviction que les pratiques de gestion généralement perçues positivement par les employés le sont également dans ce contexte.

Rainfall Climatology over Middle East Region and its Variability

A better understanding of the rainfall climatology of the Middle East region identifying the mechanisms responsible for the rain producing systems is essential for effective utilization of the water resources over the arid region. A comprehensive analysis on the rainfall climatology of the Middle East region is carried out to bring out the spatial and temporal variation of rainfall and mechanisms responsible for the rain events. The study was carried out utilizing rainfall, OLR, wind and humidity data sets procured from TRMM, NOAA and NCEP-NCAR. Climatology of annual rainfall brings out two areas of alarmingly low rainfall in the Middle East region: one in Egypt, Jordan and adjoining areas and the other in the southern part of Saudi Arabia. Daily rainfall analysis indicates that northern region gets rainfall mainly during winter and spring associated with the passage of Mediterranean low pressure systems whereas rain over the southern region is caused mainly by the monsoon organized convection, cross equatorial flow and remnants of low pressure systems associated with the monsoon during the summer season. Thermodynamic structure of the atmosphere reveals that the region does not have frequent local convection due to insufficient moisture content. The sinking motion associated with the sub tropic high pressure system and subsidence associated with the Walker circulation are responsible for maintaining warm and dry air over the region.

A Comparative Study on Internationalization and the Role of International Offices at Selected Middle Eastern and German Universities

Internationalization of higher education has become one of the most important policies for institutions of higher education worldwide. Though universities are international by nature, the need for intensified quality activities of international nature has promoted internationalization to be under spotlight of researchers, administrators and policy makers and to be an area for research. Each institution follows its certain way to govern its international affairs. Most Universities, especially in the 'Developed World' started to plan it strategically. This study explores the meanings and importance of internationalization especially that it means different things to different people. It also studies the rationales behind internationalizing higher education. It focuses on the four main prevailing rationales; political, cultural/social, economic/financial, and academic on both national and institutional levels. With the increasing need to strategically plan, the study explores internationalization strategies in terms of how to develop them, what are their approaches and types, and their components and dimensions. Damascus University has witnessed an overwhelming development of its international relations and activities. Therefore, it started to face a problem of how to deal with this increasing load especially that its International Office is the only unit that deals with the international issues. In order to study the internationalization phenomenon at Damascus University, the 2WH approach, which asks the what, why, and how questions, is used and in order to define the International Office's role in the internationalization process of the University, it studies it and the international offices of Kassel University, and Humboldt University in Germany, The University of Jordan, and Al Baath University in Syria using the 'SOCIAL' approach that studies and analyses the situation, organization, challenges, involvement, ambitions, and limitations of these offices. The internationalization process at the above-mentioned Universities is studied and compared in terms of its meaning, rationales for both the institution and its academic staff, challenges and strategic planning. Then a comparison is made among the international offices of the Universities to identify their approaches, what led to their success and what led to their failure in their practices. The aim is to provide Damascus University and its International Office with some good practices and, depending on the experiences of the professionals of the case-studies, a suggested guidance to the work of this Office and the University in general is given. The study uses the interviews with the different officials and stakeholders of the case-studies as the main method of collecting the information in addition to site visits, studying their official documents and their websites. The study belongs to qualitative research that has an action dimension in it since the recommendations will be applied in the International Office. The study concludes with few learned lessons for Damascus University and its International Office depending on the comparison that was done according to a set of dimensions. Finally a reflection on the relationship between internationalization of higher education and politics, the impact of politics on Middle Eastern Universities, and institutional internationalization strategies are presented.

Triangulation by Continuous Embedding

When triangulating a belief network we aim to obtain a junction tree of minimum state space. Searching for the optimal triangulation can be cast as a search over all the permutations of the network's vaeriables. Our approach is to embed the discrete set of permutations in a convex continuous domain D. By suitably extending the cost function over D and solving the continous nonlinear optimization task we hope to obtain a good triangulation with respect to the aformentioned cost. In this paper we introduce an upper bound to the total junction tree weight as the cost function. The appropriatedness of this choice is discussed and explored by simulations. Then we present two ways of embedding the new objective function into continuous domains and show that they perform well compared to the best known heuristic.

Probabilistic Independence Networks for Hidden Markov Probability Models

Graphical techniques for modeling the dependencies of randomvariables have been explored in a variety of different areas includingstatistics, statistical physics, artificial intelligence, speech recognition, image processing, and genetics.Formalisms for manipulating these models have been developedrelatively independently in these research communities. In this paper weexplore hidden Markov models (HMMs) and related structures within the general framework of probabilistic independencenetworks (PINs). The paper contains a self-contained review of the basic principles of PINs.It is shown that the well-known forward-backward (F-B) and Viterbialgorithms for HMMs are special cases of more general inference algorithms forarbitrary PINs. Furthermore, the existence of inference and estimationalgorithms for more general graphical models provides a set of analysistools for HMM practitioners who wish to explore a richer class of HMMstructures.Examples of relatively complex models to handle sensorfusion and coarticulationin speech recognitionare introduced and treated within the graphical model framework toillustrate the advantages of the general approach.

Neural Networks

We present an overview of current research on artificial neural networks, emphasizing a statistical perspective. We view neural networks as parameterized graphs that make probabilistic assumptions about data, and view learning algorithms as methods for finding parameter values that look probable in the light of the data. We discuss basic issues in representation and learning, and treat some of the practical issues that arise in fitting networks to data. We also discuss links between neural networks and the general formalism of graphical models.

Factorial Hidden Markov Models

We present a framework for learning in hidden Markov models with distributed state representations. Within this framework, we derive a learning algorithm based on the Expectation--Maximization (EM) procedure for maximum likelihood estimation. Analogous to the standard Baum-Welch update rules, the M-step of our algorithm is exact and can be solved analytically. However, due to the combinatorial nature of the hidden state representation, the exact E-step is intractable. A simple and tractable mean field approximation is derived. Empirical results on a set of problems suggest that both the mean field approximation and Gibbs sampling are viable alternatives to the computationally expensive exact algorithm.

Fast Learning by Bounding Likelihoods in Sigmoid Type Belief Networks

Sigmoid type belief networks, a class of probabilistic neural networks, provide a natural framework for compactly representing probabilistic information in a variety of unsupervised and supervised learning problems. Often the parameters used in these networks need to be learned from examples. Unfortunately, estimating the parameters via exact probabilistic calculations (i.e, the EM-algorithm) is intractable even for networks with fairly small numbers of hidden units. We propose to avoid the infeasibility of the E step by bounding likelihoods instead of computing them exactly. We introduce extended and complementary representations for these networks and show that the estimation of the network parameters can be made fast (reduced to quadratic optimization) by performing the estimation in either of the alternative domains. The complementary networks can be used for continuous density estimation as well.

Active Learning with Statistical Models

For many types of learners one can compute the statistically 'optimal' way to select data. We review how these techniques have been used with feedforward neural networks. We then show how the same principles may be used to select data for two alternative, statistically-based learning architectures: mixtures of Gaussians and locally weighted regression. While the techniques for neural networks are expensive and approximate, the techniques for mixtures of Gaussians and locally weighted regression are both efficient and accurate.

On Convergence Properties of the EM Algorithm for Gaussian Mixtures

"Expectation-Maximization'' (EM) algorithm and gradient-based approaches for maximum likelihood learning of finite Gaussian mixtures. We show that the EM step in parameter space is obtained from the gradient via a projection matrix $P$, and we provide an explicit expression for the matrix. We then analyze the convergence of EM in terms of special properties of $P$ and provide new results analyzing the effect that $P$ has on the likelihood surface. Based on these mathematical results, we present a comparative discussion of the advantages and disadvantages of EM and other algorithms for the learning of Gaussian mixture models.

Learning from Incomplete Data

Real-world learning tasks often involve high-dimensional data sets with complex patterns of missing features. In this paper we review the problem of learning from incomplete data from two statistical perspectives---the likelihood-based and the Bayesian. The goal is two-fold: to place current neural network approaches to missing data within a statistical framework, and to describe a set of algorithms, derived from the likelihood-based framework, that handle clustering, classification, and function approximation from incomplete data in a principled and efficient manner. These algorithms are based on mixture modeling and make two distinct appeals to the Expectation-Maximization (EM) principle (Dempster, Laird, and Rubin 1977)---both for the estimation of mixture components and for coping with the missing data.

On the Convergence of Stochastic Iterative Dynamic Programming Algorithms

Recent developments in the area of reinforcement learning have yielded a number of new algorithms for the prediction and control of Markovian environments. These algorithms, including the TD(lambda) algorithm of Sutton (1988) and the Q-learning algorithm of Watkins (1989), can be motivated heuristically as approximations to dynamic programming (DP). In this paper we provide a rigorous proof of convergence of these DP-based learning algorithms by relating them to the powerful techniques of stochastic approximation theory via a new convergence theorem. The theorem establishes a general class of convergent algorithms to which both TD(lambda) and Q-learning belong.