Boolean functions play an important role in the design and analysis of symmetric-key cryptosystems, as well as having applications in other fields such as coding theory. Boolean functions acting on large number of inputs introduces the problem of computing the cryptographic properties. Traditional methods of computing these properties involve transformations which require computation and memory resources exponential in the number of input variables. When the number of inputs is large, Boolean functions are usually defined by the algebraic normal form (ANF) representation. In this thesis, methods for computing the weight and nonlinearity of Boolean functions from the ANF representation are investigated. The relation between the ANF coefficients and the weight of a Boolean function was introduced by Carlet and Guillot. This expression allows the weight to be computed in $mathcal{O}(2^p)$ operations for a Boolean function containing $p$ monomials in its ANF. In this work, a more efficient algorithm for computing the weight is proposed, which eliminates the unnecessary calculations in the weight expression. By generalizing the weight expression, a formulation of the distances to the set of linear functions is obtained. Using this formulation, the problem of computing the nonlinearity of a Boolean function from its ANF is reduced to an associated binary integer programming problem. This approach allows the computation of nonlinearity for Boolean functions with high number of input variables and consisting of small number of monomials in a reasonable time.

Background: Schizophrenia first appears in adolescence, in boys at an earlier age than girls. The interpretation of this key epidemiological finding crucially depends on whether similar age-related sex differences exist in the expression of associated, subclinical psychosis-like experiences. Methods: Findings are based on a population sample of 2548 adolescents and young adults aged 17–28. Subjects were assessed with the core psychosis sections on delusions and hallucinations of the Munich- Composite International Diagnostic Interview. Results: The risk of subclinical psychotic experiences was significantly higher for males in the younger half of the cohort (17–21 years), but similar in the older half (22–28 years). Conclusions: These findings suggest that normal maturational changes in adolescence with differential age of onset in boys and girls cause the expression of psychosis, the extreme of which is schizophrenia.

In this thesis, Genetic Granular Cognitive Fuzzy Neural Networks (GGCFNN), combining genetic algorithms (GA) and granular cognitive fuzzy neural networks (GCFNN), is proposed for pattern recognition problems. According to cognitive patterns, biological neural networks in the human brain can recognize different patterns. Since GA and neural networks represent two learning methods based on biological science, it is indispensable and valuable to investigate how biological neural networks and artificial neural networks recognize different patterns. The new GGCFNN, based on granular computing, soft computing and cognitive science, is used in the pattern recognition problems. The hybrid forward-wave-backward-wave learning algorithm, as a main learning technology in GCFNN, is used to enhance learning quality. GA optimizes parameters to make GGCFNN get better learning results. Both pattern recognition results generated by human persons and those by GGCFNN are analyzed in terms of computer science and cognitive science.

Recombinant Bovine Somatotropin (rBST) is a biotechnology for increasing milk production in dairy cattle. The purpose of this research was to investigate and to build a better understanding of the complexities and controversies around this product in Canada. To accomplish this, I examined the Standing Senate Committee on Agriculture and Forestrys inquiry into rBST and the drug approval process. I compared and contrasted the testimony of witnesses and Senators and I uncovered emerging issues, patterns, and themes. This research was an exploratory and qualitative exercise that analyzed how the participants of this Senate inquiry conceptualized and contested the meaning of science, safety, and the states regulatory functions. <p> This research revealed several commonalities between Health Canada management, the human safety panel, and industry representatives. These witnesses argued that the drug approval process must be efficient, standard-driven, and based upon available scientific studies. These witnesses stated that they had confidence in the neutrality and competency of internal standard setting-agencies. They emphasized transparency rather than public participation in the drug approval process. Health and safety were conceptualized as static phenomena to be measured and evaluated by experts. <p>In contrast, Health Canada employees had several commonalities with the Senators, dairy representatives, and witnesses from citizen interest groups. Their testimony supports the argument that health and safety are dynamic social constructs. These actors transformed the boundaries of science to accommodate their precautionary framing of safety. They highlighted several problems with Canadas science-based regulatory framework and demanded that they have a decisive voice in the rBST decision. They challenged the hegemony of industrial capitalism by combining both scientific and lay knowledge to expose the limits and contradictions of industrialized agriculture.

<p>Risk factors such as intrauterine and vaginal infection put pregnant women at risk for delivering preterm. Bacterial vaginosis (BV) is a polymicrobial clinical syndrome commonly diagnosed in women of reproductive age, with women of African descent with low socioeconomic status and previous preterm delivery at high risk. Although frequently isolated from healthy women,&nbsp / Gardnerella vaginalis has been most frequently associated with BV. There is limited data available on the prevalence of BV in Southern Africa / therefore, we embarked on a study to determine the&nbsp / prevalence of BV and G. vaginalis in predominantly black communities in the Western Cape, in order to establish the role of G. vaginalis in BV. Women attending various Maternity and Obstetrics&nbsp / units (MOU) in the Cape Peninsula with and without a history of pre-term delivery (PTD) were invited to participate in the study. Several factors were statistically associated with pregnancy history,&nbsp / including location of study population, parity, smoking and presence of clinical symptoms. The presence of G. vaginalis was determined by culture in 51.7% of the preterm delivery group (PTDG)&nbsp / and 44% of the full-term delivery group (FTDG) women. BV was detected in 31.13% of PTDG and 23.67% of FTDG by Gram stained analysis according to Nugent scoring criteria, with age and HIV&nbsp / status posing as risk factors. When comparing PTDG and FTDG for an association between the presence of G. vaginalis and BV, a stronger association was observed in the PTDG but it was not statistically significant. In both PTDG and FTDG, G. vaginalis was isolated significantly more often in women diagnosed with BV at 24.5% (p &lt / 0.05). Antibiogram studies revealed both Metronidazole and Clindamycin resistant strains of G. vaginalis. G. vaginalis Biotype 7 is specifically associated with BV, while Biotype 2 appears to be associated with BV in women with a history&nbsp / of PTD. Accuracy of diagnostic tools were tested and it was determined that Nugent scoring is more sensitive in diagnosing BV (76.04%), but culture for G. vaginalis is more specific (83.21%). Although this study was limited in that we were unable to follow-up pregnancy outcomes, we were able to confirm the perceived role of G. vaginalis in BV.&nbsp / </p>

La tesis se centra en el estudio de las características y procedimientos de inferencia de los Modelos de Localización y Escala. Por una parte, se caracteriza a todos los modelos de localización simétricos para los cuales una combinación lineal de la media y mediana muestrales es un estimador asintóticamente eficiente del parámetro de localización. El modelo resultante, a tres parámetros, puede entenderse como una distribución Normal Truncada Simetrizada. Se presentan además dos métodos alternativos para estimar los parámetros que, por sus propiedades asintóticas, resultan buenos competidores de los estimadores de máxima verosimilitud (EMV): uno basado en la "Curtosis Empírica" que se destaca por su sencillez de cálculo y un "Algoritmo" iterativo que puede implementarse fácilmente usando software estándar que trabaje con la distribución Normal Truncada simple. También se realizan estudios basados en simulaciones a fin de comparar el comportamiento de los distintos estimadores cuando el tamaño muestral es pequeño. Extendiendo este resultado al caso particular del estimador de Hodges-Lehmann, se caracteriza a las distribución Logística como el único modelo de localización simétrico para el cual este estimador es asintóticamente eficiente. Por otra parte, se investigan los procedimientos de inferencia en modelos de localización y escala en presencia de censura de Tipo I (time censored) y se demuestra una condición suficiente para la unicidad del EMV. Además, se aplica el estadístico Z* de Barndorff-Nielsen, basado en la aproximación asintótica de orden superior Saddlepoint, para la estimación por intervalos de la media poblacional de la distribución Normal y el parámetro de escala de la distribución Valor Extremo (Log-Weibull); asimismo, y a través de simulaciones, se estudia su comportamiento para pequeñas muestras. En la extensión al caso de dos muestras se considera la comparación de medias para muestras emparejadas e independientes (problema de Behrens-Fisher) y la comparación de los parámetros de escala de dos distribuciones Valor Extremo.Si bien el ámbito de la investigación se desarrolla dentro de la Estadística Matemática, todos los tópicos tratados se ilustran con ejemplos de aplicación a situaciones prácticas. / The thesis is focused on the study of the characteristics and procedures of statistical inference for the Location and Scale Models.First, we have characterized all the symmetric location models for which a linear combination of the median and sample mean is an asymptotically efficient estimator of the location parameter. The resulting model can be understood as a symmetrized or doubled truncated normal distribution. Two alternative methods to estimate the parameters are given, which are good competitors of the maximum likelihood estimators because their asymptotic properties: one is based on the empirical curtosis and is remarkable for its simplicity; the other is a simple iterative algorithm that can be performed by using standard software working with the singly truncated normal distribution. In order to compare the available estimators we have studied their performance for small sample sizes by Monte Carlo simulations.On the other hand, for location and scale models with Type I Censored data, the estimation of the parameters based on likelihood is analysed and a sufficient condition for the uniquely of the maximum likelihood estimator is given. We have used the Barndorff-Nielsen Z*-statistic based on higher-order asymptotic methods (saddlepoint approximations) for several inferential purposes: to obtain confidence intervals for the theoretical mean of a normal distribution and for the scale parameter of an Extreme Value (Log-Weibull) distribution, to compare the means of two normal populations when the samples are matched and independents with not equally variances (Bhrens-Fisher problem) and also, to compare the scale parameters of two Extreme Value distribution. More over, the performance of the Z*-statistic for small samples sizes is investigated by Monte Carlo experiments. Although the investigation is basically developed on the field of Mathematical Statistics, examples of applications to practical situations are given illustrate all the topics considered.

Un sous-groupe unique d'individus obèses nommés, «métaboliquement en santé, mais obèses» (MHO) a été identifié dans la littérature médicale. Malgré leur obésité, ces individus semblent être protégés du développement de perturbations métaboliques ce qui pourrait diminuer leurs risques de développer le diabète de type 2 ainsi que des maladies cardiovasculaires (Meigs et al., 2006). L'objectif de ce projet était de déterminer si le captage et la mise en réserve des acides gras libres (AGL) pourraient expliquer ce profil métabolique favorable. Plus précisément, le but était de déterminer l'activité et l'expression de la lipoprotéine lipase (LPL) ainsi que l'expression du diaglycérol acyltranférase (DGAT1) et du transporteur membranaire «Cluster of differenciation 36» (CD36) chez deux groupes de femmes obèses post-ménopausées, soient les MHO et les «à risque», c'est-à-dire présentant des complications métaboliques. L'hypothèse était que les individus MHO auraient des activités et des expressions plus hautes de LPL et du DGAT1 ainsi qu'une plus grande expression du CD36 membranaire que les individus «à risque». Deux sous-groupes de femmes post-ménopausées étaient à l'étude : les individus MHO (n=15) et obèses «à risque» (n=30). Suite à un test d'hyperglycémie provoquée par voie orale, les volontaires étaient classées dans les deux groupes selon l'index Matsuda, à partir de la sensibilité à l'insuline. Les données mesurées incluent un profil lipidique complet, un test glycémique et insulinémique sur 2h ainsi qu'une biopsie du tissu adipeux abdominal, afin de déterminer les activités et expressions de la LPL, du DGAT1 et du CD36. Aucune différence significative n'a été observée entre les individus MHO et «à risque» pour l'activité de la LPL ainsi que pour l'ARNm du DGAT1, du CD36 et de la LPL. Toutefois, une tendance favorisant les individus MHO a été observée concernant l'ARNm du CD36 comparé aux individus «à risque» (1,01 ± 0,4 vs 0,82 ± 0,2 respectivement, P=0,07). De plus, les résultats de la régression linéaire indiquent que l'ARNm du CD36 est une variable qui explique 7% de la variance de la sensibilité à l'insuline. Les résultats de cette étude préliminaire indiquent que le captage des AGL, plus spécifiquement l'ARNm du CD36, peut être un mécanisme fondamental qui pourrait expliquer le profil métabolique favorable des individus MHO. Des études plus approfondies devront être effectuées avec un plus grand nombre de sujets afin de mieux caractériser les individus MHO au point de vue fondamental. ______________________________________________________________________________ MOTS-CLÉS DE L’AUTEUR : MHO, obésité, acides gras libres, maladies cardiovasculaires

És difícil dibuixar una frontera, dins la Teoria de Sistemas Dinàmics, entre lleis de conservació i simetries doncs, sovint, les seves característiques es confonen. Un clar exemple d'aquest fenómen el constitueixen els sistemes Hamiltonians i els sistemes reversibles.Breument, un sistema dinàmic es diu temps-reversible (o, per nosaltres, simplement reversible) si és invariant sota l'acció d'un difeomorfisme involutiu a l'espai i una inversió en el sentit del temps. és en aquest marc on cal situar aquesta memòria. Concretament, ens centrem en dos punts molt particulars: la Teoria de Formes Normals i el fenómen del trencament de separatrius, tots dos introduïts per Poincaré a la seva tesi (1890).Respecte al primer d'aquests punts, en aquesta tesi s'introdueix el concepte de Pseudo Forma Normal (breument PNF), inspirat en idees d'en Moser, i que permet transformar, sota certes hipotesis, un sistema analític en un d'equivalent d'aspecte el més simple possible. Aquesta PNF és una generalització de la coneguda Forma Normal de Birkhoff amb la qual coincideix si el sistema considerat és Hamiltonià o reversible. Com a conseqüència, s'obté, en determinats casos, l'equivalència local entre aquests dos tipus de sistemes. Aquesta PNF pot esdevenir una eina útil per estudiar la dinàmica d'un sistema analític a l'entorn d'un equilibri (un punt, una òrbita periòdica o un tor).El segon punt, l'escissió de separatrius, fa referència a l'intersecció transversal de varietats invariants procedent del trencament d'una certa connexió homoclínica a l'afegir al sistema una petita pertorbació. Un dels motius d'interés sobre aquest fenòmen és que és un dels principals causants de comportament estocàstic en sistemes Hamiltonians.Un problema relacionat amb aquest trencament de separatrius és el de mesurar-lo, sigui a partir del càlcul de l'angle amb el que es troben aquestes varietats per primer cop, per l'àrea que tanquen entre elles, etc. El mètode habitualment utilitzat per a estimar-lo és l'anomenat mètode de Poincaré-Melnikov. Malhauradament, si la pertorbació és ràpidament oscil-latòria els termes que proporciona aquest mètode són exponencialment petits en el paràmetre pertorbador, fet que dificulta el seu càlcul. En aquesta tesi s'ha demostrat, tal i com passa en el cas Hamiltonià, que en el cas d'un sistema reversible, respecte a una involució lineal, 2-dimensional i pertorbat de manera ràpidament periòdica i reversible, el mètode de Poincaré-Melnikov és correcte i dóna en primer ordre l'anomenada funció de Melnikov. / It is difficult, in the Theory of Dynamical Systems, to draw a boundary line between conservation laws and symmetries because often their effects on the dynamics are very similar. This is the case of the Hamiltonian and the Reversible systems.Briefly, a dynamical system is called time-reversible (or, simply, reversible) if it is invariant under the action of an involutive spatial diffeomorphism and a reversion in time's arrow. This is the frame where this work must be placed. Precisely, we focus our attention in two particular points: the Theory of Normal Forms and the phenomenon of the splitting of separatrices, both introduced by Poincare in his thesis (1890).Regarding the first one of this topics, we introduce the concept of Pseudo-normal Form (PNF in short). It comes from ideas of Moser and allows to transform, under suitable conditions, an analytic system around an equilibrium in another equivalent one having a quite simple form. This PNF is a generalization of the celebrated Birkhoff Normal Form and both coincide if the system is Hamiltonian or reversible. Consequently, the local equivalence between both types of systems is derived in some cases. This PNF can become a useful tool to study the dynamics of an analytic system in a neighborhood of an equilibrium (a fixed point, a periodic orbit or a torus).The second topic, the splitting of separatrices, is related to the transversal intersection of invariant manifolds derived from the splitting of a given homoclinic connection when some small perturbation is considered. One of the reasons that makes this phenomenon interesting is that it seems to be one of the main causes of the stochastic behavior in Hamiltonian systems.One problem related to this splitting of separatrices becomes to measure it, studying, for instance, some angle the form when they meet for the first time, the area of the first lobe, etc. The standard method to estimate this size is the celebrated Poincaré-Melnikov method. Unfortunately, if the perturbation oscillates rapidly the terms provided by this method are exponentially small in the perturbation parameter, and this fact makes this computation more involved. In this work we prove, like it happens in the Hamiltonian case, that in the case of a 2-dimensional analytic reversible system (reversible with respect to a linear spatial involution) perturbed by a rapidly periodic reversible perturbation, the Poincaré-Melnikov method works and it provides, at first order, the well known Melnikov Function.

We study numerical stability for interior-point methods applied to Linear Programming, LP, and Semidefinite Programming, SDP. We analyze the difficulties inherent in current methods and present robust algorithms. <br /><br /> We start with the error bound analysis of the search directions for the normal equation approach for LP. Our error analysis explains the surprising fact that the ill-conditioning is not a significant problem for the normal equation system. We also explain why most of the popular LP solvers have a default stop tolerance of only 10^-8 when the machine precision on a 32-bit computer is approximately 10^-16. <br /><br /> We then propose a simple alternative approach for the normal equation based interior-point method. This approach has better numerical stability than the normal equation based method. Although, our approach is not competitive in terms of CPU time for the NETLIB problem set, we do obtain higher accuracy. In addition, we obtain significantly smaller CPU times compared to the normal equation based direct solver, when we solve well-conditioned, huge, and sparse problems by using our iterative based linear solver. Additional techniques discussed are: crossover; purification step; and no backtracking. <br /><br /> Finally, we present an algorithm to construct SDP problem instances with prescribed strict complementarity gaps. We then introduce two <em>measures of strict complementarity gaps</em>. We empirically show that: (i) these measures can be evaluated accurately; (ii) the size of the strict complementarity gaps correlate well with the number of iteration for the SDPT3 solver, as well as with the local asymptotic convergence rate; and (iii) large strict complementarity gaps, coupled with the failure of Slater's condition, correlate well with loss of accuracy in the solutions. In addition, the numerical tests show that there is no correlation between the strict complementarity gaps and the geometrical measure used in [31], or with Renegar's condition number.

L'objectif de ce travail de recherche consistait à utiliser la réalité virtuelle (simulateur de conduite) afin d'améliorer les capacités attentionnelles de personnes présentant une baisse d'efficience de ces aptitudes (en lien avec l'âge ou la survenue d'un traumatisme crânien) pouvant se répercuter sur la conduite automobile. En référence au modèle de l'attention de Van Zomeren et Brouwer (1994), quatre processus attentionnels (alerte phasique, vigilance, attention sélective et attention divisée) ont été entraînés sur simulateur de conduite lors d'un programme comprenant 9 sessions d'une heure. Un groupe contrôle a effectué un training plus conventionnel sur ordinateur. Les performances des participants (10 patients traumatisés crâniens et 16 personnes âgées de plus de 65 ans) ont été évaluées avant et après le training à partir de trois mesures : une évaluation neuropsychologique, une évaluation sur simulateur de conduite et une évaluation sur route réelle. Le bénéfice de l'entraînement sur les performances aux tests neuropsychologiques s'est avéré limité pour les deux groupes. Concernant l'évaluation sur simulateur, les résultats ont révélé une amélioration des performances plus importante pour les participants entraînés sur simulateur. En revanche, les participants ayant bénéficié du training sur ordinateur ont davantage progressé que ceux entraînés sur simulateur lors de l'évaluation sur route réelle. Les résultats ont été discutés en termes d'approche rééducative et d'automatisation des processus attentionnels. Ce travail permet de conclure quant à la nécessité de poursuivre et d'approfondir les recherches sur la réalité virtuelle et le transfert de l'apprentissage en situation de vie quotidienne.