The thesis deals with the figure of Lady Wisdom in the Old Testament in the wisdom books of Job, Proverbs, Wisdom and Sirach. The opening part specifies the selected biblical texts of the Old Testament, the concept of wisdom in antiquity from Egyptian and Mesopotamian sources, and their possible inspiration for authors of wise scriptures. Next the thesis briefly introduces the chosen biblical texts and analyses the concept of wisdom in the each of them. To illustrate the nature and conduct of Wisdom the next part describes the importace of building a house for the ancient people. As an opposite to Lady Wisdom there is Lady Foolishness. Final part describes various forms of personified Wisdom in particular biblical books explaining her transformation within the course of Old Testament history.

The energy consumption to make the off-chip memory writing and readings are aknown problem. In the image processing field structure from motion simpler compressiontechniques could be used to save energy. A balance between the detected features suchas corners, edges, etc., and the degree of compression becomes a big issue to investigate.In this thesis a deeper study of this balance are performed. A number of more advancedcompression algorithms for processing of still images such as JPEG is used for comparisonwith a selected number of simpler compression algorithms. The simpler algorithms canbe divided into two categories: individual block-wise compression of each image andcompression with respect to all pixels in each image. In this study the image sequences arein grayscale and provided from an earlier study about rolling shutters. Synthetic data setsfrom a further study about optical flow is also included to see how reliable the other datasets are. / Energikonsumtionen för att skriva och läsa till off-chip minne är ett känt problem. Inombildbehandlingsområdet struktur från rörelse kan enklare kompressionstekniker användasför att spara energi. En avvägning mellan detekterade features såsom hörn, kanter, etc.och grad av kompression blir då en fråga att utreda. I detta examensarbete har en djuparestudie av denna avvägning utförts. Ett antal mer avancerade kompressionsalgoritmer förbearbetning av stillbilder som tex. JPEG används för jämförelse med ett antal utvaldaenklare kompressionsalgoritmer. De enklare algoritmerna kan delas in i två kategorier:individuell blockvis kompression av vardera bilden och kompression med hänsyn tillsamtliga pixlar i vardera bilden. I studien är bildsekvenserna i gråskala och tillhandahållnafrån en tidigare studie om rullande slutare. Syntetiska data set från ytterligare en studie om’optical flow’ ingår även för att se hur pass tillförlitliga de andra dataseten är.

Dans le De constancia sapientis, Sénèque formule le portrait d'un sage que l'injustice et l'outrage n'affectent pas. Selon le Commentaire au livre de la Sagesse de Robert Holkot (OP + 1349), cette description conceptuelle définit la notion de sagesse présente dans le livre biblique. Ainsi, dans les années 1336-1338, le sage stoïcien est-il présenté, dans une exégèse biblique, comme l'exemple même de sagesse. Partant, la thèse s'attache à retracer l'élaboration de cette figure sapientielle au prisme du prologue de ce commentaire biblique. Elle en analyse ensuite l'enjeu conceptuel : la revendication d'un modèle vie chrétien inspiré du comportement exemplaire des philosophes antiques. Dans ce contexte, la figure de Platon acquiert une importance cruciale. Une approche critique de la philosophie comme modèle de vie forme enfin la dernière partie de ce travail. En effet, tant le recours antique et médiéval à l'exemplum que la définition de la philosophie comme manière de vivre prennent appui sur une conception de l'acte moral comme imitation d'un héros (saint, sage, etc.) qu'Immanuel Kant congédie. / In the De constancia sapientis, Seneca portrays a wise man as unaffected by injustice and outrage. For Robert Holkot (O.P., †1349), writing in his Commentary on the Book of Wisdom, this conceptual description defines the notion of wisdom present in the Biblical book of the same name. Thus, in the years 1336-1338, the Stoic wise man appears in a work of Biblical exegesis as the very example of wisdom. This thesis takes as its point of departure the portrayal of the wise man in the prologue to the Commentary on the Book of Wisdom. It then analyses what is at stake: the assertion of a Christian model of life inspired by role models from ancient philosophy. Plato becomes a figure of crucial importance. Finally, the last part of this work consists in a criticism of philosophy as a model way of life. In fact, both the ancient and medieval use of exempla and the definition of philosophy as way of life depend on a conception of moral action as the imitation of a hero (saint, wise person, and so on), a conception dismissed by Immanuel Kant.

[pt] A crescente disponibilidade de dados de sensoriamento remoto vem criando novas oportunidades e desafios em aplicações de monitoramento de processos naturais e antropogénicos em escala global. Nos últimos anos, as técnicas de aprendizado profundo tornaram-se o estado da arte na análise de dados de sensoriamento remoto devido sobretudo à sua capacidade de aprender automaticamente atributos discriminativos a partir de grandes volumes de dados. Um dos problemas chave em análise de imagens é a segmentação semântica, também conhecida como rotulação de pixels. Trata-se de atribuir uma classe a cada sítio de imagem. As chamadas redes totalmente convolucionais de prestam a esta função. Os anos recentes têm testemunhado inúmeras propostas de arquiteturas de redes totalmente convolucionais que têm sido adaptadas para a segmentação de dados de observação da Terra. O presente trabalho avalias cinco arquiteturas de redes totalmente convolucionais que representam o estado da arte em segmentação semântica de imagens de sensoriamento remoto. A avaliação considera dados provenientes de diferentes plataformas: veículos aéreos não tripulados, aeronaves e satélites. Cada um destes dados refere-se a aplicações diferentes: segmentação de espécie arbórea, segmentação de telhados e desmatamento. O desempenho das redes é avaliado experimentalmente em termos de acurácia e da carga computacional associada. O estudo também avalia os benefícios da utilização do Campos Aleatórios Condicionais (CRF) como etapa de pósprocessamento para melhorar a acurácia dos mapas de segmentação. / [en] The increasing availability of remote sensing data has created new opportunities and challenges for monitoring natural and anthropogenic processes on a global scale. In recent years, deep learning techniques have become state of the art in remote sensing data analysis, mainly due to their ability to learn discriminative attributes from large volumes of data automatically. One of the critical problems in image analysis is the semantic segmentation, also known as pixel labeling. It involves assigning a class to each image site. The so-called fully convolutional networks are specifically designed for this task. Recent years have witnessed numerous proposals for fully convolutional network architectures that have been adapted for the segmentation of Earth observation data. The present work evaluates five fully convolutional network architectures that represent the state of the art in semantic segmentation of remote sensing images. The assessment considers data from different platforms: unmanned aerial vehicles, airplanes, and satellites. Three applications are addressed: segmentation of tree species, segmentation of roofs, and deforestation. The performance of the networks is evaluated experimentally in terms of accuracy and the associated computational load. The study also assesses the benefits of using Conditional Random Fields (CRF) as a post-processing step to improve the accuracy of segmentation maps.

Les modélisateurs rencontrent souvent des incertitudes sur la manière de concevoir un modèle logiciel particulier. Les recherches existantes ont montré comment les modélisateurs peuvent travailler en présence de ce type d' ''incertitude au moment de la conception''. Cependant, le processus par lequel les développeurs en viennent à exprimer leurs incertitudes reste flou. Dans cette thèse, nous prenons des pas pour combler cette lacune en proposant de créer un langage de modélisation d'incertitude et une approche pour articuler l'incertitude au moment de la conception. Nous illustrons notre proposition sur un exemple et l'évaluons non seulement sur deux scénarios d'ingénierie logicielle, mais aussi sur une étude de cas réel basée sur les incertitudes causées par la pandémie COVID-19. Nous menons également un questionnaire post-étude avec les chercheurs qui ont participé à l'étude de cas. Afin de prouver la faisabilité de notre approche, nous fournissons deux outils et les discutons. Enfin, nous soulignons les avantages et discutons des limites de notre travail actuel. / Modellers often encounter uncertainty about how to design a particular software model. Existing research has shown how modellers can work in the presence of this type of ''design-time uncertainty''. However, the process by which developers come to elicit and express their uncertainties remains unclear. In this thesis, we take steps to address this gap by proposing to create an uncertainty modelling language and an approach for articulating design-time uncertainty. We illustrate our proposal on a worked example and evaluate it not only on two software engineering scenarios, but also on a real case study based on uncertainties caused by the COVID-19 pandemic. We also conduct a post-study questionnaire with the researchers who participated in the case study. In order to prove the feasibility of our approach, we provide two tool supports and discuss them. Finally, we highlight the benefits and discuss the limitations of our current work.

The performance and lifetime of lithium-ion batteries are strongly influenced by their composition. One category of critical components are electrolyte additives, which are included primarily to stabilize electrode/electrolyte interfaces in the battery cells by forming passivation layers. The presented study aimed to identify and study such an additive that could form a hydrogen-evolution-suppressing solid electrolyte interphase (SEI) in lithium-ion batteries based on aqueous electrolytes. A promising molecular additive, ethyl 2,2-difluoroacetate (EDFA), was found to hold the qualities required for an SEI former and was herein further analyzed electrochemically. Analysis of the battery cells were performed with linear sweep voltammetry and cyclic voltammetry with varying scan rate and EDFA concentrations. Results show that both 1 and 10 w-% EDFA in the electrolyte produced hydrogen-evolution-suppressing SEI:s, although the higher concentration provided no apparent benefit. Lithium-ion full-cells based on LiMn2O4 vs. Li4Ti5O12 active materials displayed poor, though partly reversible, dis-/charge cycling despite the operation of the electrode far outside the electrochemical stability window of the electrolyte. Inclusion of reference electrodes in the lithium-ion cells proved to be immensely challenging with unpredictable drifts in their electrode potentials during operation. To summarize, HER-suppressing electrolyte additives are demonstrated to be a promising approach to stabilize high-voltage operation of aqueous lithium-ion cells although further studies are necessary before any practical application thereof can be realized. Electrochemical evaluation of the reaction mechanism and efficiency of the electrolyte additives relies however heavily on the use of reference electrodes and further development thereof is necessary.

Indiana University-Purdue University Indianapolis (IUPUI) / This thesis proposes an artificial neural network (NN) model to predict fuel consumption in heavy vehicles. The model uses predictors derived from vehicle speed, mass, and road grade. These variables are readily available from telematics devices that are becoming an integral part of connected vehicles. The model predictors are aggregated over a fixed distance traveled (i.e., window) instead of fixed time interval. It was found that 1km windows is most appropriate for the vocations studied in this thesis. Two vocations were studied, refuse and delivery trucks. The proposed NN model was compared to two traditional models. The first is a parametric model similar to one found in the literature. The second is a linear regression model that uses the same features developed for the NN model. The confidence level of the models using these three methods were calculated in order to evaluate the models variances. It was found that the NN models produce lower point-wise error. However, the stability of the models are not as high as regression models. In order to improve the variance of the NN models, an ensemble based on the average of 5-fold models was created. Finally, the confidence level of each model is analyzed in order to understand how much error is expected from each model. The mean training error was used to correct the ensemble predictions for five K-Fold models. The ensemble K-fold model predictions are more reliable than the single NN and has lower confidence interval than both the parametric and regression models.

No description available.

La médecine personnalisée promet des soins adaptés à chaque patient. Cependant, l’ap- prentissage automatique appliqué à cette fin nécessite beaucoup d’améliorations. L’évalua- tion des modèles est une étape cruciale qui nécessite du travail pour amener à un niveau acceptable pour son utilisation avec des participants. Actuellement, les performances sur les ensembles de données biomédicales sont évaluées à l’aide d’un découpage intra-sujet ou inter-sujet. Le premier se concentre sur l’évaluation des participants présents à la fois dans les ensembles d’entraînement et de test. Ce dernier sépare les participants pour chaque ensemble. Ces termes sont respectivement synonymes de fractionnement aléatoire et par sujet. Deux méthodes principales se présentent comme des solutions pour obtenir des performances de franctionnement aléatoires lors d’entraînement de méthodes par sujet, calibration et sans ca- libration. Alors que la calibration se concentre sur l’entraînement d’un petit sous-ensemble de participant non vues, les méthodes sans calibration visent à modifier l’architecture du modèle ou les traitements préliminaire pour contourner la nécessité du sous-ensemble. Ce mémoire étudiera la calibration non paramétrique pour ses propriétés d’indépendance de la modalité. L’article présenté détaillera cette enquête pour combler l’écart de performance sur un ensemble de données d’essais de marche sur des surfaces irrégulières. Nous détermi- nons que quelques cycles (1-2) de marche sont suffisants pour calibrer les modèles pour des performances adéquates (F1 : +90%). Avec accès à des essais de cycle de marche supplémen- taires (+10), le modèle a atteint à peu près les mêmes performances qu’un modèle formé à l’aide d’une approche de fractionnement aléatoire (F1 : 95-100%). Suivant les objectifs de la médecine personnalisée, des voies de recherche supplémentaires sont décrites, telles qu’une méthode alternative de distribution de modèles qui s’adapte aux étapes de recherche tout en réduisant les coûts de calcul pour les développeurs de modèles. Nous constatons que l’étalonnage est une méthode valable pour surmonter l’écart de performance. Les ré- sultats correspondent aux découvertes précédentes utilisant l’étalonnage pour obtenir des performances robustes. / Personalized medicine promises care tailored to each patient; however, machine learning applied to this end needs much improvement. Evaluation of models is a crucial step which necessitates attention when utilized with participants. Currently, performance on biomedical datasets is evaluated using either intra-subject or inter-subject splitting. The former focuses on the evaluation of participants present in both training and testing sets. The latter separates participants for each set. These terms are synonymous with random-wise and subject-wise splitting, respectively. Two main methods present themselves as solutions to achieving random-wise performance while training on a subject-wise dataset split, calibration and calibration-free methods. While calibration focuses on training a small subset of unseen data trials, calibration-free methods aim to alter model architecture or pre-processing steps to bypass the necessity of training data points. This thesis investigates non-parametric calibration for its modality-agnostic properties. The article presented details this investigation at bridging the performance gap on a dataset of gait trials on irregular surfaces. We determine few (1-2) gait cycles are sufficient to calibrate models for adequate performance (F1:+90%). With access to additional gait cycle trials, the model achieved nearly the same performance as a model trained using a random-split approach (F1:95-100%). Following the goals of personalized medicine, additional research paths are outlined, such as an alternative model distribution method which fits with research steps while reducing computational costs for model developers. We find that calibration is a valid method to overcome the performance gap. The presented results correspond with previous findings by using calibration to achieve robust performance.

No description available.