abstract: This work focuses on a generalized assessment of source zone natural attenuation (SZNA) at chlorinated aliphatic hydrocarbon (CAH) impacted sites. Given the numbers of sites and technical challenges for cleanup there is a need for a SZNA method at CAH impacted sites. The method anticipates that decision makers will be interested in the following questions: 1-Is SZNA occurring and what processes contribute? 2-What are the current SZNA rates? 3-What are the longer-term implications? The approach is macroscopic and uses multiple lines-of-evidence. An in-depth application of the generalized non-site specific method over multiple site events, with sampling refinement approaches applied for improving SZNA estimates, at three CAH impacted sites is presented with a focus on discharge rates for four events over approximately three years (Site 1:2.9, 8.4, 4.9, 2.8kg/yr as PCE, Site 2:1.6, 2.2, 1.7, 1.1kg/y as PCE, Site 3:570, 590, 250, 240kg/y as TCE). When applying the generalized CAH-SZNA method, it is likely that different practitioners will not sample a site similarly, especially regarding sampling density on a groundwater transect. Calculation of SZNA rates is affected by contaminant spatial variability with reference to transect sampling intervals and density with variations in either resulting in different mass discharge estimates. The effects on discharge estimates from varied sampling densities and spacings were examined to develop heuristic sampling guidelines with practical site sampling densities; the guidelines aim to reduce the variability in discharge estimates due to different sampling approaches and to improve confidence in SZNA rates allowing decision-makers to place the rates in perspective and determine a course of action based on remedial goals. Finally bench scale testing was used to address longer term questions; specifically the nature and extent of source architecture. A rapid in-situ disturbance method was developed using a bench-scale apparatus. The approach allows for rapid identification of the presence of DNAPL using several common pilot scale technologies (ISCO, air-sparging, water-injection) and can identify relevant source architectural features (ganglia, pools, dissolved source). Understanding of source architecture and identification of DNAPL containing regions greatly enhances site conceptualization models, improving estimated time frames for SZNA, and possibly improving design of remedial systems. / Dissertation/Thesis / Ph.D. Civil and Environmental Engineering 2013

CoordenaÃÃo de AperfeiÃoamento de NÃvel Superior / Este trabalho propÃe melhorias em uma tÃcnica de amostragem adaptativa multidimensional para renderizaÃÃo. RenderizaÃÃo à o processo de sÃntese de imagens por meio de algoritmos que simulam a iluminaÃÃo em cenÃrios virtuais. As tÃcnicas mais gerais de renderizaÃÃo fotorrealÃstica â aquelas que procuram obter imagens que se assemelham a fotografias â utilizam mÃtodos de integraÃÃo baseados em Monte Carlo para resolver a equaÃÃo que descreve a distribuiÃÃo de luz na cena (equaÃÃo de renderizaÃÃo). Por ser um mÃtodo probabilÃstico e utilizar amostras geradas randomicamente, Monte Carlo produz ruÃdo na imagem final â resultado da variÃncia das amostras â e portanto, pode necessitar de uma grande quantidade de amostras para que o ruÃdo diminua a nÃveis aceitÃveis. Com o intuito de se obter imagens de melhor qualidade com uma menor quantidade de amostras, foram pospostas tÃcnicas de amostragem adaptativa que visam concentrar o esforÃo de amostragem em regiÃes mais importantes da cena. Neste trabalho, propÃe-se a modificaÃÃo de uma tÃcnica de amostragem adaptativa multidimensional por meio da adiÃÃo de duas etapas: substituiÃÃo de amostras e integraÃÃo auxiliar. Essas etapas visam dar mais robustez à tÃcnica, possibilitando sua utilizaÃÃo em uma maior variedade de situaÃÃes. AlÃm da adiÃÃo de duas etapas, tambÃm propÃe-se uma tÃcnica de reconstruÃÃo mais eficiente na etapa final. / This work proposes improvements in a multidimensional adaptive sampling technique for rendering. Rendering is the process of synthesizing images by algorithms simulating lighting in virtual scenes. The more general techniques of photorealistic rendering â those seeking images that resemble photographs â use integration methods based on Monte Carlo to solve the equation that describes the distribution of light in the scene (rendering equation). Being a probabilistic method which uses randomly generated samples, Monte Carlo produces noise in the final image â result of samplesâ variance â and therefore may require a large amount of samples to reduce the noise to acceptable levels. To obtain images of better quality with a lower number of samples, adaptive sampling techniques were proposed, concentrating sampling effort in the most important regions. In this work, we propose the addition of two steps to a multidimensional adaptive sampling technique: substitution of samples and auxiliary integration. These steps aim to give more strength to the technique, enabling their use in a wider variety of situations.

In this thesis several sampling methods for Statistical Learning with imbalanced data have been implemented and evaluated with a new metric, imbalanced accuracy. Several modifications and new algorithms have been proposed for intelligent sampling: Border links, Clean Border Undersampling, One-Sided Undersampling Modified, DBSCAN Undersampling, Class Adjusted Jittering, Hierarchical Cluster Based Oversampling, DBSCAN Oversampling, Fitted Distribution Oversampling, Random Linear Combinations Oversampling, Center Repulsion Oversampling. A set of requirements on a satisfactory performance metric for imbalanced learning have been formulated and a new metric for evaluating classification performance has been developed accordingly. The new metric is based on a combination of the worst class accuracy and geometric mean. In the testing framework nonparametric Friedman's test and post hoc Nemenyi’s test have been used to assess the performance of classifiers, sampling algorithms, combinations of classifiers and sampling algorithms on several data sets. A new approach of detecting algorithms with dominating and dominated performance has been proposed with a new way of visualizing the results in a network. From experiments on simulated and several real data sets we conclude that: i) different classifiers are not equally sensitive to sampling algorithms, ii) sampling algorithms have different performance within specific classifiers, iii) oversampling algorithms perform better than undersampling algorithms, iv) Random Oversampling and Random Undersampling outperform many well-known sampling algorithms, v) our proposed algorithms Hierarchical Cluster Based Oversampling, DBSCAN Oversampling with FDO, and Class Adjusted Jittering perform much better than other algorithms, vi) a few good combinations of a classifier and sampling algorithm may boost classification performance, while a few bad combinations may spoil the performance, but the majority of combinations are not significantly different in performance.

Cette thèse est une contribution à la problématique de la simulation d'événements rares. A partir de l'étude des méthodes de Splitting, un nouveau cadre théorique est développé, indépendant de tout algorithme. Ce cadre, basé sur la définition d'un processus ponctuel associé à toute variable aléatoire réelle, permet de définir des estimateurs de probabilités, quantiles et moments sans aucune hypothèse sur la variable aléatoire. Le caractère artificiel du Splitting (sélection de seuils) disparaît et l'estimateur de la probabilité de dépasser un seuil est en fait un estimateur de la fonction de répartition jusqu'au seuil considéré. De plus, les estimateurs sont basés sur des processus ponctuels indépendants et identiquement distribués et permettent donc l'utilisation de machine de calcul massivement parallèle. Des algorithmes pratiques sont ainsi également proposés.Enfin l'utilisation de métamodèles est parfois nécessaire à cause d'un temps de calcul toujours trop important. Le cas de la modélisation par processus aléatoire est abordé. L'approche par processus ponctuel permet une estimation simplifiée de l'espérance et de la variance conditionnelles de la variable aléaoire résultante et définit un nouveau critère d'enrichissement SUR adapté aux événements rares / This thesis address the issue of extreme event simulation. From a original understanding of the Splitting methods, a new theoretical framework is proposed, regardless of any algorithm. This framework is based on a point process associated with any real-valued random variable and lets defined probability, quantile and moment estimators without any hypothesis on this random variable. The artificial selection of threshold in Splitting vanishes and the estimator of the probability of exceeding a threshold is indeed an estimator of the whole cumulative distribution function until the given threshold. These estimators are based on the simulation of independent and identically distributed replicas of the point process. So they allow for the use of massively parallel computer cluster. Suitable practical algorithms are thus proposed.Finally it can happen that these advanced statistics still require too much samples. In this context the computer code is considered as a random process with known distribution. The point process framework lets handle this additional source of uncertainty and estimate easily the conditional expectation and variance of the resulting random variable. It also defines new SUR enrichment criteria designed for extreme event probability estimation.

L’information quantique peut être traitée sur différents types de systèmes physiques. Elle peut également être traitée selon deux façons fondamentalement différentes, en usant soit des variables discrètes, soit des variables continues. Dans cette thèse, nous nous concentrons sur l’optique quantique en variables continues et les expériences étudiées sont basées sur l’utilisation de photons. Après une introduction à quelques notions de base de la mécanique quantique, nous présentons un protocole sous-universel d’informatique quantique, appelé Boson Sampling, précédé d’une partie exposant des éléments de théorie de la complexité nécessaires pour comprendre la preuve de supériorité quantique de ce protocole. Puis, nous proposons un modèle pour décrire la création de qudits intriqués dans une expérience réalisée au sein de notre équipe. Enfin, la dernière partie de cette thèse présente une interprétation physique plus fondamentale des résultats obtenus lors d’expériences de type Hong-Ou-Mandel avec des filtres en fréquence devant les photodétecteurs, et montre qu’il s’agit d’une partie d’un état chat de Schrödinger produit par post-sélection. / Quantum information can be processed on differents types of physical systems. It can also be processed in two fundamentally different ways, using either discrete or continuous variable implementations. In this thesis, we concentrate on quantum optics in continuous variables and the studied experiments are based on the use of photons. After an introduction to some basic notions of quantum mechanics, we present a subuniversal protocol of quantum computing, named Boson Sampling, preceded by a part exposing elements of complexity theory which are necessary to understand the quantum superiority proof of this protocol. Then, we propose a model to describe the creation of entangled qudits in an experiment done in the team. Finally, the last part of this thesis presents a more fundamental physical interpretation of the results obtained during Hong-Ou-Mandel experiments with frequency filters in front of the photodetectors, and shows that we have a part of a Schrödinger cat state, produced by postselection.

This thesis explores the attitudes of women with borderline personality disorder towards otherness. The aim of the thesis was to explore the relationship between different personality characteristics of five women diagnosed with borderline personality disorder and their attitudes towards themselves and representatives of otherness. For these purposes, the Semi- Structured Interview for Personality Functioning (STiP-5.1) assessing personal functioning in the personal and interpersonal domains and the experience sampling method (ESM) were used, through which the respondents recorded the intensity of emotions felt towards themselves and various representatives of otherness for 30 days. The data were processed qualitatively into five case studies. The results indicated significant intraindividual variability in emotions toward self and representatives of otherness. The methods could find application in the treatment process and therapy, whereby assessment of personality functioning could contribute to the overall current picture of the patient and the variability of their psychological attitudes in response to certain life events could be monitored using ESM. KEYWORDS borderline personality disorder, otherness, STiP-5.1, experience sampling methodology

No description available.

The objective of this dissertation is to investigate GLR control charts based on a sequential sampling scheme (SS GLR charts). Phase II monitoring is considered and the goal is to quickly detect a wide range of changes in the univariate normal process mean parameter and/or the variance parameter. The performance of the SS GLR charts is evaluated and design guidelines for SS GLR charts are provided so that practitioners can easily apply the SS GLR charts in applications. More specifically, the structure of this dissertation is as follows: We first develop a two-sided SS GLR chart for monitoring the mean μ of a normal process. The performance of the SS GLR chart is evaluated and compared with other control charts. The SS GLR chart has much better performance than that of the fixed sampling rate GLR chart. It is also shown that the overall performance of the SS GLR chart is better than that of the variable sampling interval (VSI) GLR chart and the variable sampling rate (VSR) CUSUM chart. The SS GLR chart has the additional advantage that it requires fewer parameters to be specified than other VSR charts. The optimal parameter choices are given, and regression equations are provided to find the limits for the SS GLR chart. If detecting one-sided shifts in μ is of interest, the above SS GLR chart can be modified to be a one-sided chart. The performance of this modified SS GLR chart is investigated. Next we develop an SS GLR chart for simultaneously monitoring the mean μ and the variance 𝜎² of a normal process. The performance and properties of this chart are evaluated. The design methodology and some illustrative examples are provided so that the SS GLR chart can be easily used in applications. The optimal parameter choices are given, and the performance of the SS GLR chart remains very good as long as the parameter choices are not too far away from the optimized choices. / Ph. D.

No description available.

No description available.