Made available in DSpace on 2016-06-02T20:25:20Z (GMT). No. of bitstreams: 1 5382.pdf: 2589978 bytes, checksum: 40194cb079db8b86aaa133e8e75657ef (MD5) Previous issue date: 2013-06-07 / In this research we realize an investigation into the assessment process in a course of English Language using a thematic task-based planning for teachers pre-service of adverse context of a private college in the state of São Paulo, and check if there or not the compatibility of the assessment instruments and activities of teaching material - "A Short Historical Sketch of Britain" (Barbirato, 2005) with such planning principles, criteria and tasks definitions. Being aware that the evaluation process should understand the context researched in evaluative instruments that are consistent with the instruction given in the classroom, we list the purpose of this study to ascertain the extent to which these instruments measure facilitated the learning of FL through a theme, seeking the meaning of language and communication as the main unease in this area is to find a way to evaluate the student consistent with such instruction. We choose a qualitative interpretative in nature, through the gathering of evaluative data documents used throughout the course, as well as a questionnaire for teacher s course who contributed to the triangulation of the data analyzed. The results allowed us to infer that the evaluative instruments, although some still aren t the most appropriate measure progress achieved before the student's ability to communicate at English Language. However, analysis of the data documents revealed that although some instruments, especially the written evaluations, require changes to be in line with course design model implemented one the planning in question, can say that it contributed to reaching a positive evaluation in this context. / Nesta pesquisa realizamos uma investigação sobre o processo avaliativo em um curso de LE utilizando um planejamento temático baseado em tarefas para professores préserviço de contexto adverso de uma Faculdade Particular do interior do estado de São Paulo, além de verificar se houve ou não a compatibilidade dos instrumentos avaliativos e das atividades do material didático - A Short Historical Sketch of Britain (BARBIRATO,2005) com os princípios de tal planejamento, dos critérios e definições de tarefa. Tendo consciência de que o processo avaliativo no contexto pesquisado deve compreender em instrumentos avaliativos que se coadunam com a instrução dada em sala de aula, elencamos o objetivo deste trabalho de verificar em que medida tais instrumentos facilitaram aferir a aprendizagem da LE por meio de um tema, visando o significado da língua e a comunicação, pois a principal inquietação nesta área é encontrar uma maneira de avaliar o aluno condizente com tal instrução. Optamos por uma pesquisa qualitativa de natureza interpretativista, por meio do recolhimento dos documentos de dados avaliativos utilizados no decorrer do curso, bem como a aplicação de um questionário para a professora ministrante que colaborou para a triangulação dos dados analisados. Os resultados obtidos permitiram-nos inferir que os instrumentos avaliativos, apesar de alguns ainda não serem os mais adequados, conseguiram aferir a evolução do aluno diante da capacidade de se comunicar em Língua Estrangeira. No entanto, a análise dos documentos de dados revelou que apesar de alguns instrumentos, principalmente as avaliações escritas, precisarem de algumas modificações para estar em consonância com o planejamento em questão, podemos dizer que contribuíram para chegar a uma possível avaliação adequada a esse contexto.

The problem discussed in this thesis is the optimization of the process of planning, tracking and evaluating tasks in the management area of the department Corporate Technology in Siemens, Brno. Having studied theories of CMMI and principles of ISO 9001, the given process is optimized by automating certain manual tasks. As Corporate Technology uses a SharePoint server to manage tasks, the application is developed to improve his functionality. This application has therefore been tested in the environment of the organization with real-life data and complies to all necessary requirements to be able to save at least 50 % of time in the optimized process. The results of this thesis prove the usability of automating manual repetitive tasks being conducted in the processes of the organization and introduce possibility of increasing process level by preventing human errors.

L'intelligence artificielle (IA) moderne a ouvert de nouvelles perspectives prometteuses pour la création de robots intelligents. En particulier, les architectures d'apprentissage basées sur le gradient (réseaux neuronaux profonds) ont considérablement amélioré la compréhension des scènes 3D en termes de perception, de raisonnement et d'action. Cependant, ces progrès ont affaibli l'attrait de nombreuses techniques ``classiques'' développées au cours des dernières décennies. Nous postulons qu'un mélange de méthodes ``classiques'' et ``apprises'' est la voie la plus prometteuse pour développer des modèles du monde flexibles, interprétables et exploitables : une nécessité pour les agents intelligents incorporés. La question centrale de cette thèse est : ``Quelle est la manière idéale de combiner les techniques classiques avec des architectures d'apprentissage basées sur le gradient pour une compréhension riche du monde 3D ?''. Cette vision ouvre la voie à une multitude d'applications qui ont un impact fondamental sur la façon dont les agents physiques perçoivent et interagissent avec leur environnement. Cette thèse, appelée ``programmes différentiables pour modèler l'environnement'', unifie les efforts de plusieurs domaines étroitement liés mais actuellement disjoints, notamment la robotique, la vision par ordinateur, l'infographie et l'IA. Ma première contribution---gradSLAM--- est un système de localisation et de cartographie simultanées (SLAM) dense et entièrement différentiable. En permettant le calcul du gradient à travers des composants autrement non différentiables tels que l'optimisation non linéaire par moindres carrés, le raycasting, l'odométrie visuelle et la cartographie dense, gradSLAM ouvre de nouvelles voies pour intégrer la reconstruction 3D classique et l'apprentissage profond. Ma deuxième contribution - taskography - propose une sparsification conditionnée par la tâche de grandes scènes 3D encodées sous forme de graphes de scènes 3D. Cela permet aux planificateurs classiques d'égaler (et de surpasser) les planificateurs de pointe basés sur l'apprentissage en concentrant le calcul sur les attributs de la scène pertinents pour la tâche. Ma troisième et dernière contribution---gradSim--- est un simulateur entièrement différentiable qui combine des moteurs physiques et graphiques différentiables pour permettre l'estimation des paramètres physiques et le contrôle visuomoteur, uniquement à partir de vidéos ou d'une image fixe. / Modern artificial intelligence (AI) has created exciting new opportunities for building intelligent robots. In particular, gradient-based learning architectures (deep neural networks) have tremendously improved 3D scene understanding in terms of perception, reasoning, and action. However, these advancements have undermined many ``classical'' techniques developed over the last few decades. We postulate that a blend of ``classical'' and ``learned'' methods is the most promising path to developing flexible, interpretable, and actionable models of the world: a necessity for intelligent embodied agents. ``What is the ideal way to combine classical techniques with gradient-based learning architectures for a rich understanding of the 3D world?'' is the central question in this dissertation. This understanding enables a multitude of applications that fundamentally impact how embodied agents perceive and interact with their environment. This dissertation, dubbed ``differentiable world programs'', unifies efforts from multiple closely-related but currently-disjoint fields including robotics, computer vision, computer graphics, and AI. Our first contribution---gradSLAM---is a fully differentiable dense simultaneous localization and mapping (SLAM) system. By enabling gradient computation through otherwise non-differentiable components such as nonlinear least squares optimization, ray casting, visual odometry, and dense mapping, gradSLAM opens up new avenues for integrating classical 3D reconstruction and deep learning. Our second contribution---taskography---proposes a task-conditioned sparsification of large 3D scenes encoded as 3D scene graphs. This enables classical planners to match (and surpass) state-of-the-art learning-based planners by focusing computation on task-relevant scene attributes. Our third and final contribution---gradSim---is a fully differentiable simulator that composes differentiable physics and graphics engines to enable physical parameter estimation and visuomotor control, solely from videos or a still image.