L’approche au coeur de cette étude est une recherche-action qui a pour but de mettre à l’essai un réseau littéraire, à caractère interdisciplinaire, pour le réinvestissement du vocabulaire disciplinaire et des concepts en arithmétique auprès des élèves de deuxième année du primaire. Cette recherche comporte trois objectifs spécifiques : 1) mettre à l’essai un réseau littéraire pour décrire et analyser les pratiques enseignantes 2) décrire et analyser le réinvestissement du vocabulaire disciplinaire en arithmétique des élèves du premier cycle du primaire 3) mettre en relation les pratiques enseignantes et le réinvestissement du vocabulaire disciplinaire en arithmétique des élèves. Pour y arriver, une enseignante de 2e année du primaire du centre de services scolaire Marguerite-Bourgeoys a participé à notre recherche en expérimentant le réseau littéraire proposé dans cette étude. Nous avons conduit trois séances d’entretien en tout avec l’enseignante avant, pendant et après l’expérimentation. Une analyse de contenu a été effectuée à partir des propos et des impressions de l’enseignante. Durant l’expérimentation, nous avons participé à trois séances d’observation dont l’analyse a été effectuée en nous basant sur le modèle multi-agenda. Finalement, quatre élèves ont participé à un entretien qui visait notamment à analyser et à décrire le réinvestissement du vocabulaire disciplinaire en arithmétique et leurs résultats ont été analysés à l’aide d’une grille d’évaluation bâtie par nous. À la lumière des données, nous constatons l’importance de l’engagement professionnel et des pratiques expertes de l’enseignante pour favoriser le réinvestissement du vocabulaire disciplinaire associé aux concepts en arithmétique chez les élèves. Concernant les apports du réseau littéraire, les albums jeunesse ont constitué une source de motivation, de participation et d’engagement pour les élèves. Ils ont aussi favorisé le réinvestissement du vocabulaire disciplinaire. Les limites de cette recherche se situent davantage au niveau de la longueur et du sens de certaines questions étant donné que certains élèves n’ont pas le français comme langue maternelle, ce qui pouvait avoir un impact sur le niveau de compréhension des questions en lien avec les oeuvres du réseau. / This research project fits according to an action research approach and aims at putting into practice a literacy network, being of interdisciplinary nature, for the reinvestment of disciplinary vocabulary in arithmetic with second grade students. This present research has the following three specific objectives: 1) to put into practice a literacy network to analyze and describe the teacher’s practices 2) to analyze and describe the second-grade student’s reinvestment of disciplinary vocabulary in arithmetic 3) to put into relation the impact of teacher practices in the reinvestment of the student’s disciplinary vocabulary in arithmetic. In order to succeed, a second-grade elementary teacher from the Centre de services scolaire Marguerite-Bourgeoys participated in our research project by experimenting the literacy network proposed in this study. During the experimentation, we participated in three sessions of observation in which the analysis was performed according to the multi-agenda model. In addition, we conducted three semi-structured interview sessions before, during and after the experimentation. A content analysis was carried out based on the teacher’s comments and impressions. Lastly, four students participated in our interview questionnaire and their results were analyzed by an evaluation grid that we have built. At the light of the results and of our analysis, we notice the importance of professional engagement and the second-grade teacher’s expert practices pertaining to the disciplinary vocabulary of arithmetic of the students. Concerning the contributions, children’s literature has constituted an important source of student motivation and participation. It also favored the reinvestment of disciplinary vocabulary. The limits of this research are more pertaining to the length and the meaning of certain questions considering the fact that certain students do not have French as their native language which could’ve had an impact on their level of comprehension relating to the questions on the literacy network.

This study lays a foundation for the empirical mapping of the material production of Afrikaans fiction from 1990 to 2005. The primary research question, namely, how to map Afrikaans fiction production during this period, has been answered through both descriptive-qualitative and statistical-quantitative empirical research. In the first, contextual descriptive component, qualitative research is used to describe the context within which Afrikaans fiction production takes place. The impact of transformation shifts on the production and producer landscapes are mapped through a description of the publishing system. Within this system, book production is regarded as an institution that has inter-systemic relationships with other sub-systems, including institutional contexts (such as media, cultural, literary, educational and library sub-systems), as well as broader societal contexts (such as political, economic and technological contexts). Transformation shifts in the producer landscape (including the establishment, conglomeration, mergers, take-overs and closures of publishing houses) are described through profiling the relevant producers (including publishing houses and other kinds of producers) who contribute to this landscape. In the second, statistical empirical component, quantitative research is used to sketch the contours of Afrikaans fiction production through the construction of production and producer profiles. On a theoretical level, the study follows a sociological and institutional approach to empirical literary studies. This study has been positioned within the developing field of book and publishing studies, specifically within the parameters of the epistemology of production statistics. Itamar Even-Zohar’s poly-system theory and Pierre Bourdieu’s theory of the field of cultural production form the theoretical framework. Several methodological instruments were developed for the purpose of this study, including a new model of the publishing system, a database of production statistics (the Production Database of Afrikaans Fiction – PDAF), and two typologies of kinds of producers and fiction publications according to which both producer and production categories can be distinguished. The publishing system, the PDAF and the two typologies have been proven useful instruments for the mapping of book production. These instruments will contribute to the fields of book and publishing studies through their application to research on other production categories. The PDAF is a useful resource for measuring the growth and diversity of Afrikaans fiction production. The study concludes that Afrikaans fiction production between 1990 and 2005 was incisively influenced by transformation shifts. Publishing in this field became more challenging in that Afrikaans fiction production had to wean itself from the artificial institutional privilege and substantial state support it had enjoyed and was forced to become economically independent. By 2005, there were far fewer significant producers of Afrikaans fiction in the producer landscape than in 1990. In some production categories, fiction production has shrunk, while, in others, it has grown. Overall, the production profile of Afrikaans fiction has greatly diversified, with Afrikaans fiction appearing in a wide variety of production categories. Afrikaans fiction remains the most diverse and best-developed indigenous tradition of fiction production. The challenge for future case studies is to investigate the relationships between production patterns and specific transformation shifts in the publishing system in terms of specific production categories. / Thesis (PhD)--University of Pretoria, 2008. / Information Science / unrestricted

<div>The rise of artificial intelligence (AI) hinges on the efficient training of modern deep neural networks (DNNs) for non-convex optimization and uncertainty quantification, which boils down to a non-convex Bayesian learning problem. A standard tool to handle the problem is Langevin Monte Carlo, which proposes to approximate the posterior distribution with theoretical guarantees. However, non-convex Bayesian learning in real big data applications can be arbitrarily slow and often fails to capture the uncertainty or informative modes given a limited time. As a result, advanced techniques are still required.</div><div><br></div><div>In this thesis, we start with the replica exchange Langevin Monte Carlo (also known as parallel tempering), which is a Markov jump process that proposes appropriate swaps between exploration and exploitation to achieve accelerations. However, the na\"ive extension of swaps to big data problems leads to a large bias, and the bias-corrected swaps are required. Such a mechanism leads to few effective swaps and insignificant accelerations. To alleviate this issue, we first propose a control variates method to reduce the variance of noisy energy estimators and show a potential to accelerate the exponential convergence. We also present the population-chain replica exchange and propose a generalized deterministic even-odd scheme to track the non-reversibility and obtain an optimal round trip rate. Further approximations are conducted based on stochastic gradient descents, which yield a user-friendly nature for large-scale uncertainty approximation tasks without much tuning costs. </div><div><br></div><div>In the second part of the thesis, we study scalable dynamic importance sampling algorithms based on stochastic approximation. Traditional dynamic importance sampling algorithms have achieved successes in bioinformatics and statistical physics, however, the lack of scalability has greatly limited their extensions to big data applications. To handle this scalability issue, we resolve the vanishing gradient problem and propose two dynamic importance sampling algorithms based on stochastic gradient Langevin dynamics. Theoretically, we establish the stability condition for the underlying ordinary differential equation (ODE) system and guarantee the asymptotic convergence of the latent variable to the desired fixed point. Interestingly, such a result still holds given non-convex energy landscapes. In addition, we also propose a pleasingly parallel version of such algorithms with interacting latent variables. We show that the interacting algorithm can be theoretically more efficient than the single-chain alternative with an equivalent computational budget.</div>

For many years, community colleges that chose to offer community college baccalaureate (CCB) programs were looked upon in a negative light (Rice, 2015). However, as the need for specialized baccalaureates within specific fields and job markets have continued to grow (McKee, 2005), CCB programs are becoming more widely accepted throughout the United States. In spite of this paradigm shift, Mississippi is one of the remaining states that have not embraced the idea of CCB programs, in spite of its statistical deficiency in regards to baccalaureate degree holding citizens (Williams, 2010). The focus of this study was to explore the perceptions of community college administrators in Mississippi with regards to the influence of CCB programs to the community college mission of institutions in their state. This study indicates that administrators in Mississippi recognize the benefits of offering CCB programs, but do not want CCB programs to take away from the well-established statewide higher education system through mission creep. Many of the strong position statements received overwhelmingly neutral responses. In contrast, Administrators who chose to give their opinion indicated that they are not familiar with how CCB programs are implemented, and do not believe that Mississippi is ready for CCB programs across the state. However, respondents felt that the community college mission is always evolving, should meet students’ needs, and varies from location to location. The findings also show that administrators are favorable to the piloting of CCB programs at a few (1-2) institutions, even though they believe the programs will take funding away from current programs and do not want community colleges evolving into 4-year institutions. The study also concludes that there is a significant difference between institution size and survey questions #18 and #20. There is also a significant difference between length of time in the community college sector and survey questions #15, #17, and #18.

SLOVENE NATIONAL THEATRE IN LJUBLJANA Author Report Of The Diploma Work Author: Bc. Petr Hýl Supervisor: doc. ing. arch. Zdeněk Makovský

Text in Afrikaans / Psychology / D.Litt. et Phil (Psychology)

Text in Afrikaans / Psychology / D.Litt. et Phil (Psychology)