Ce manuscrit décrit le processus optique non linéaire de Génération de Second Harmonique (acronyme anglais SHG pour Second Harmonic Generation) réalisé en phase liquide. En particulier, la propriété de cohérence de ce processus est étudiée en détail. En effet, en raison de la parité du processus SHG, cette cohérence est perdue dans les liquides. Ces études portent ainsi sur plusieurs géométries afin d’accéder aux réponses de volume et de surface. Les avantages d’une méthode combinée sont aussi discutés. Dans une première partie, la configuration classique de la Diffusion Hyper Rayleigh (acronyme anglais HRS pour Hyper Rayleigh Scattering ou SHS Second harmonic scattering) est utilisée pour l’étude de systèmes simples comme les solvants purs. Cette étude porte en particulier sur les méthodes de normalisation des hyperpolarisabilités moléculaires. Une voie alternative est proposée basée sur l’introduction d’une section efficace HRS ou SHS. Par la suite, les effets liés aux ordres non linéaires supérieurs sont démontrés et interprétés, ceux-ci pouvant altérer les valeurs absolues qui peuvent être mesurées.La seconde partie porte sur l’introduction de plusieurs améliorations dans les mesures HRS ou SHS. En particulier, l’évolution vers des géométries non conventionnelles s’éloignant de la géométrie standard ainsi que sur des évolutions permettant d’accéder aux aspects dynamiques sont proposées. Des expériences sont réalisées sur des systèmes simples comme les solvants purs, des composés moléculaires et des nanoparticules, tous présentant des spécificités propres. Cette partie finit sur l’exploration des systèmes aléatoires diffusants et les problèmes associés. Enfin, dans une dernière partie, la réponse HRS ou SHS de solutions aqueuses de sels est discutée. Les mesures portent plus particulièrement sur une étude comparée des aspects cohérents et incohérents et les propriétés qu’il est possible de mesurer dans ces systèmes / This manuscript describes the nonlinear optical process of Second Harmonic Generation (SHG acronym for Second Harmonic Generation) carried out in liquid phase. In particular, this process coherence property is studied in detail. Indeed, due to the parity of the SHG process, this coherence is lost in liquids. These studies covers several geometries to access the volume and surface responses. The benefits of a combined method are also discussed. In the first part, the typical configuration of the Hyper Rayleigh Scattering(HRS or SHS for Second harmonic scattering) is used for the study of simple systems like pure solvents. This particular study focuses on methods of standardization of molecular hyperpolarizabilities. An alternative route is proposed based on the introduction of a HRS or SHS cross. Subsequently, the effects of the higher nonlinear orders are demonstrated and interpreted, they can alter the absolute values that can be measured.The second part deals with the introduction of several improvements in the HRS or SHS measurement. In particular, the trend towards unconventional geometries away from the standard geometry as well as developments for accessing dynamic aspects are proposed. Experiments are performed on simple systems such as pure solvents, molecular compounds and the nanoparticles, all presenting specificities. This part ends on exploring the random scattering systems and associated problems. Finally, in the last part, the HRS or SHS response of aqueous salt solutions is discussed. The measures focus specifically on a comparative study of coherent and incoherent aspects and properties that can be measured in these systems

Cílem této diplomové práce je vytvoření nástroje pro optimalizaci hyper-parametrů umělých neuronových sítí. Tento nástroj musí být schopen optimalizovat více hyper-parametrů, které mohou být navíc i korelovány. Tento problém jsem vyřešil implmentací optimalizátoru, který využívá Gaussovské procesy k predikci vlivu jednotlivých hyperparametrů na výslednou přesnost neuronové sítě. Z provedených experimentů na několika benchmark funkcích jsem zjistil, že implementovaný nástroj je schopen dosáhnout lepších výsledků než optimalizátory založené na náhodném prohledávání a snížit tak v průměru počet potřebných kroků optimalizace. Optimalizace založená na náhodném prohledávání dosáhla lepších výsledků pouze v prvních krocích optimalizace, než si optimalizátor založený na Gaussovských procesech vytvoří dostatečně přesný model problému. Nicméně téměř všechny experimenty provedené na datasetu MNIST prokázaly lepší výsledky optimalizátoru založeného na náhodném prohledávání. Tyto rozdíly v provedených experimentech jsou pravděpodobně dány složitostí zvolených benchmark funkcí nebo zvolenými parametry implementovaného optimalizátoru.

The thesis deals with hyper duplex stainless austenitic-ferritic steels and their mechanical and castability properties. The evaluation of resistance to pitting corrosion is divided by PRE values. Steels with a PRE value higher than 48 belong to the group of hyper duplex steels. The theoretical part contains an introduction to the chemical composition, structure and properties of these steels. The practical part is focused on the conditions of tendency to crack castings during solidification and cooling. Furthermore, there is evaluated influence of intermetallic phases on mechanical properties.

Afin de poursuivre et d'améliorer la tâche de détection qui est en cours à l'INSA, nous nous sommes concentrés sur la fusion des informations visibles et infrarouges du point de vue de reconnaissance des obstacles, ainsi distinguer entre les véhicules, les piétons, les cyclistes et les obstacles de fond. Les systèmes bimodaux ont été proposées pour fusionner l'information à différents niveaux: des caractéristiques, des noyaux SVM, ou de scores SVM. Ils ont été pondérés selon l'importance relative des capteurs modalité pour assurer l'adaptation (fixe ou dynamique) du système aux conditions environnementales. Pour évaluer la pertinence des caractéristiques, différentes méthodes de sélection ont été testés par un PPV, qui fut plus tard remplacée par un SVM. Une opération de recherche de modèle, réalisée par 10 fois validation croisée, fournit le noyau optimisé pour SVM. Les résultats ont prouvé que tous les systèmes bimodaux VIS-IR sont meilleurs que leurs correspondants monomodaux. / To continue and improve the detection task which is in progress at INSA laboratory, we focused on the fusion of the information provided by visible and infrared cameras from the view point of an Obstacle Recognition module, this discriminating between vehicles, pedestrians, cyclists and background obstacles. Bimodal systems have been proposed to fuse the information at different levels:of features, SVM's kernels, or SVM’s matching-scores. These were weighted according to the relative importance of the modality sensors to ensure the adaptation (fixed or dynamic) of the system to the environmental conditions. To evaluate the pertinence of the features, different features selection methods were tested by a KNN classifier, which was later replaced by a SVM. An operation of modelsearch, performed by 10 folds cross-validation, provides the optimized kernel for the SVM. The results have proven that all bimodal VIS-IR systems are better than their corresponding monomodal ones.

Multi-objective evolutionary algorithms have gained a lot of atten- tion in the recent years. They have proven to be among the best multi-objective optimizers and have been used in many industrial ap- plications. However, their usability is hindered by the large number of evaluations of the objective functions they require. These can be expensive when solving practical tasks. In order to reduce the num- ber of objective function evaluations, surrogate models can be used. These are a simple and fast approximations of the real objectives. In this work we present the results of research made between the years 2009 and 2013. We present a multi-objective evolutionary algo- rithm with aggregate surrogate model, its newer version, which also uses a surrogate model for the pre-selection of individuals. In the next part we discuss the problem of selection of a particular type of model. We show which characteristics of the various models are im- portant and desirable and provide a framework which combines sur- rogate modeling with meta-learning. Finally, in the last part, we ap- ply multi-objective optimization to the problem of hyper-parameters tuning. We show that additional objectives can make finding of good parameters for classifiers faster. 1

Le travail présenté dans ce manuscrit concerne la fabrication de fantômes d'aorte patient spécifiques utilisant une technique de fabrication additive par impression 3D. Ces répliques sont fabriquées en matériaux synthétiques dont les caractéristiques morphologiques et les propriétés mécaniques doivent être proches de celles déterminées sur un patient. Elles permettent d'optimiser ou de développer les techniques d'imagerie médicale, de comprendre les relations entre le comportement mécanique de la paroi aortique et les caractéristiques hémodynamiques du flux sanguin mais aussi de réaliser des entrainements préopératoires aux interventions chirurgicales, telles que le traitement endovasculaire. Dans cette étude, le comportement mécanique hyper-viscoélastique de la paroi aortique est modélisé par un modèle de Maxwell solide généralisé, dont les paramètres ont permis la sélection et le développement de matériaux élastomères de type silicone aux comportements mécaniques contrôlés. Ces matériaux ont été élaborés à partir de mélanges de formulations existantes et des lois de mélange ont été comparées pour guider la définition de la composition idéale permettant d'imiter le comportement mécanique désiré. Nous avons mis au point une méthode basée sur l'imagerie médicale par ultrason, capable d'identifier les paramètres hyper-viscoélastiques d'une paroi vasculaire. Cette méthode a été validée sur des tubes réalisés avec ces formulations de silicone, dont les propriétés mécaniques ont été mesurées avec des méthodes de référence. Puis, ces silicones ont été utilisés dans un processus de fabrication additive utilisant l'impression 3D par voie indirecte. Un travail de conception assistée par ordinateur a été réalisé pour produire un fantôme d'aorte patient-spécifique présentant un anévrisme fusiforme et non-thrombosé dans la région thoracique / The present work deals with the producing of patient-specific aortic phantoms using an additive manufacturing technique by 3D printing. Phantoms are manufactured from synthetic materials with morphological and mechanical characteristics which should be close to these identified on a patient. They can be used to develop techniques of medical imaging, to understand the relationship between aortic mechanical behavior and hemodynamic properties of blood flow, as well as to perform a preoperative training of interventions, such as endovascular treatment. In this study, the hyper-viscoelastic aortic mechanical behavior was described using a generalized solid Maxwell model. Silicone materials were developed based on the model’s mechanical parameters to mimic various aortic mechanical behaviors. These materials were formulated from commercials silicones, and then mixing rules were compared to define the ideal mixture which can mimic the specific mechanical behavior. A nondestructive method based on medical imaging by ultrasound was developed to identify the parameters of a blood vessel hyper-viscoelastic model. Silicone tubes made of our formulations with known reference mechanical parameters, were used to validate this method. Then, these silicone materials were used in an additive manufacturing process using indirect 3D printing. A work of computer aided design was done to produce a patient-specific aortic phantom with a thoracic fusiform aneurysm without thrombosis

People around the globe rely on their mobile devices for instant access to entertainment and social media. Children comprise a large majority of individuals who use smartphone applications, particularly for gaming and learning opportunities. Understandably, these apps become part of the identity development process, including the formation of one's gender identity. App developers include gendered content to capture and maintain children's attention, but much of the existing research examines children in late childhood and early adolescence, leaving the ages of six to eight relatively undiscussed. The researcher utilized a thematic analysis to review 20 children's mobile applications for instances of gender stereotypes. Social cognitive theory offers a guiding principle for understanding the process of developing one's gender identity, as well as the role that external stimuli, such as digital media examples and parent models, play. This study aimed to determine whether mobile applications targeting the identified age group use gender stereotypes, as well as how they employ these stereotypes within the application. The researcher randomly selected 20 top children's applications on the Apple App Store and examined them for gendered instances, such as occupations and interests as well as character depictions. The results reflect that instances of gender stereotypes do occur in the children's mobile applications. Many of the applications portrayed feminine stereotypes surrounding nurturing and caregiving tasks ("Mommy in Training"), making it one of the most frequently exploited feminine stereotypes in the sample. The "Boys will be Boys" stereotype comprised the most frequently displayed masculine stereotypes across the studied applications. These findings represent the idea that society places higher value on these stereotypes than others, such as social relationships ("The Power of Motivational Friendship") or recklessness ("The Risk Taker"). Implications include modeling of traditionally masculine and feminine stereotypes for young users by utilizing popular characters recognizable by most children in the target age range. / Master of Arts / Handheld electronic devices, such as smartphones and tablets, encompass some of the most widely used electronic devices in today's society. Most families in America have at least one mobile device with internet capability. Apple, the manufacturer of perhaps the most popular brand of electronic devices, pre-install their App Store on all devices they sell, giving users instant access to hundreds of thousands of different mobile applications that offer functions to make every aspect of life simpler. Young children spend a significant amount of their time playing games on these devices, although the American Academy of Pediatrics (2017) recommends that parents limit their children's daily screen time to no more than 2 hours, depending on the age of the child. The games that children download and play impose a number of different messages and stereotypes on their users, including gender stereotypes. Due to the substantial time children dedicate to these apps, the messages communicated regarding gender play crucial roles in the development of their gender identity. Social cognitive theory offers valuable insight and guidance into the gender identity development process. Therefore, the present study examines the gender stereotypes conveyed within 20 of the top children's mobile applications available on the Apple App Store in April 2018. The researcher randomly selected 20 children's applications, 10 each from the Top Free and Top Paid categories and examined them for gendered instances, such as occupations and interests as well as character depictions. The results reflect that instances of gender stereotypes do occur in the children's mobile applications. These represent the idea that society places higher value on certain stereotypes, like being caring and nurturing ("Mommy in Training") or engaging in messy, adventurous play ("Boys will be Boys"), than others, such as social relationships ("The Power of Motivational Friendship") or recklessness ("The Risk Taker"). Implications include modeling of traditionally masculine and feminine stereotypes for young users by utilizing popular characters recognizable by most children in the target age range.

Multiculturalism or cultural pluralism is a policy, ideal, or reality that emphasizes the unique characteristics of different cultures in the world, especially as they relate to one another in immigrant receiving nations. The word was first used in 1957 to describe Switzerland but first came into common currency in Canada in the late 1960s2. It quickly spread to other English- speaking and western developed countries.Although There is no clear link between the multiculturalism and the term that so called Balkanization, But this research tries to present the similar possibility of Multiculturalism discourse and Balkanization as a geopolitical term. In fact this research believes that   balkanization is The same idea of multiculturalism in practice, when it comes to the idea of state. On the other hand Vijay Parshad in his book Everybody was Kung Fu fighting  says that "I am interested in “how an investigation of kung fu can help us move from a limited multicultural framework into an antiracist, polycultural one.” This is the research and video that has been made about Bruce lee statue in Mostar, the divided city in Bosnia and Herzegovina", as a method of working in hyper- politicized society.

ITC/USA 2005 Conference Proceedings / The Forty-First Annual International Telemetering Conference and Technical Exhibition / October 24-27, 2005 / Riviera Hotel & Convention Center, Las Vegas, Nevada / The Hyper-X program flew X-43A research vehicles to hypersonic speeds over the Pacific Ocean in March and November 2004 from the Western Aeronautical Test Range, NASA Dryden Flight Research Center, Edwards, California. The program required multiple telemetry ground stations to provide continuous coverage of the captive carry, launch, boost, experiment, and descent phases of these missions. An overview is provided of vehicle telemetry and distributed assets that supported telemetry acquisition, best-source selection, radar tracking, video tracking, flight termination systems, and voice communications. Real-time data display and processing are discussed, and postflight analysis and comparison of data acquired are presented.

Hyper-heuristics is a new field of optimisation which has recently emerged and is receiving growing exposure in the research community and literature. Hyper-heuristics are optimisation methods which are designed with a high level of abstraction from any one specific problem or class of problems and therefore are more generally applicable than specialised meta-heuristic and heuristic methods. Instead of being designed to solve a specific real-world problem, hyper-heuristics are designed to solve the problem of heuristic generation and selection. As such, hyper-heuristics can be thought of as methods for optimising the operations of an optimisation process which finds good solutions to a problem as a by-product. This approach has been shown to be very effective and in some cases provides improvement in search performance as well as reducing the burden associated with tailoring meta-heuristics which is often required when solving new problems. In this thesis, the hypothesis that hyper-heuristics can be competitively applied to real-world multi-objective optimisation problems such as the water distribution design problem is tested. Although many single-objective hyper-heuristics have been proposed in the literature, only a few multi-objective methods have been proposed. This thesis explores two different novel multi-objective hyper-heuristics: one designed for generating new specialised heuristics; and one designed for solving the online selection of heuristics. Firstly, the behaviour of a set of heuristics is explored to create a base understanding of different heuristic behavioural traits in order to better understand the hyper-heuristic behaviours and dynamics later in the study. Both approaches are tested on a range of benchmark optimisation problems and finally applied to real-world instances of the water distribution network design problem where the selective hyper-heuristics is demonstrated as being very effective at solving this difficult problem. Furthermore, the thesis demonstrates how heuristic selection can be improved by incorporating a greater level of information about heuristic performance, namely the historical joint performance of different heuristics, and shows that exploiting this sequencing information in heuristic selection can produce highly competitive results.