Dissertations / Theses on the topic 'Causal graphs'
To see the other types of publications on this topic, follow the link: Causal graphs.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Causal graphs.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Choudhry, Arjun. "Narrative Generation to Support Causal Exploration of Directed Graphs." Thesis, Virginia Tech, 2020. http://hdl.handle.net/10919/98670.
Full textAbstract:
Causal graphs are a useful notation to represent the interplay between the actors as well
as the polarity and strength of the relationship that they share. They are used extensively
in educational, professional, and industrial contexts to simulate different scenarios, validate
behavioral aspects, visualize the connections between different processes, and explore the
adversarial effects of changing certain nodes. However, as the size of the causal graphs increase,
interpreting them also becomes increasingly tougher. In such cases, new analytical
tools are required to enhance the user's comprehension of the graph, both in terms of correctness
and speed. To this purpose, this thesis introduces 1) a system that allows for causal
exploration of directed graphs, while enabling the user to see the effect of interventions on
the target nodes, 2) the use of natural language generation techniques to create a coherent
passage explaining the propagation effects, and 3) results of an expert user study validating
the efficacy of the narratives in enhancing the user's understanding of the causal graphs. In
overall, the system aims to enhance user experience and promote further causal exploration.Master of Science
Narrative generation is the art of creating coherent snippets of text that cumulatively describe a succession of events, played across a period of time. These goals of narrative generation are also shared by causal graphs – models that encapsulate inferences between the nodes through the strength and polarity of the connecting edges. Causal graphs are an useful mechanism to visualize changes propagating amongst nodes in the system. However, as the graph starts addressing real-world actors and their interactions, it becomes increasingly difficult to understand causal inferences between distant nodes, especially if the graph is cyclic. Moreover, if the value of more than a single node is altered and the cumulative effect of the change is to be perceived on a set of target nodes, it becomes extremely difficult to the human eye. This thesis attempts to alleviate this problem by generating dynamic narratives detailing the effect of one or more interventions on one or more target nodes, incorporating time-series analysis, Wikification, and spike detection. Moreover, the narrative enhances the user's understanding of the change propagation occurring in the system. The efficacy of the narrative was further corroborated by the results of user studies, which concluded that the presence of the narrative aids the user's confidence level, correctness, and speed while exploring the causal network.
2
Bernigau, Holger. "Causal Models over Infinite Graphs and their Application to the Sensorimotor Loop." Doctoral thesis, Universitätsbibliothek Leipzig, 2015. http://nbn-resolving.de/urn:nbn:de:bsz:15-qucosa-164734.
Full textAbstract:
Motivation and background
The enormous amount of capabilities that every human learns throughout his life, is probably among the most remarkable and fascinating aspects of life. Learning has therefore drawn lots of interest from scientists working in very different fields like philosophy, biology, sociology, educational sciences, computer sciences and mathematics. This thesis focuses on the information theoretical and mathematical aspects of learning.
We are interested in the learning process of an agent (which can be for example a human, an animal, a robot, an economical institution or a state) that interacts with its environment. Common models for this interaction are Markov decision processes (MDPs) and partially observable Markov decision processes (POMDPs). Learning is then considered to be the maximization of the expectation of a predefined reward function. In order to formulate general principles (like a formal definition of curiosity-driven learning or avoidance of unpleasant situation) in a rigorous way, it might be desirable to have a theoretical framework for the optimization of more complex functionals of the underlying process law. This might include the entropy of certain sensor values or their mutual information. An optimization of the latter quantity (also known as predictive information) has been investigated intensively both theoretically and experimentally using computer simulations by N. Ay, R. Der, K Zahedi and G. Martius. In this thesis, we develop a mathematical theory for learning in the sensorimotor loop beyond expected reward maximization.
Approaches and results
This thesis covers four different topics related to the theory of learning in the sensorimotor loop.
First of all, we need to specify the model of an agent interacting with the environment, either with learning or without learning. This interaction naturally results in complex causal dependencies. Since we are interested in asymptotic properties of learning algorithms, it is necessary to consider infinite time horizons. It turns out that the well-understood theory of causal networks known from the machine learning literature is not powerful enough for our purpose. Therefore we extend important theorems on causal networks to infinite graphs and general state spaces using analytical methods from measure theoretic probability theory and the theory of discrete time stochastic processes. Furthermore, we prove a generalization of the strong Markov property from Markov processes to infinite causal networks.
Secondly, we develop a new idea for a projected stochastic constraint optimization algorithm. Generally a discrete gradient ascent algorithm can be used to generate an iterative sequence that converges to the stationary points of a given optimization problem. Whenever the optimization takes place over a compact subset of a vector space, it is possible that the iterative sequence leaves the constraint set. One possibility to cope with this problem is to project all points to the constraint set using Euclidean best-approximation. The latter is
sometimes difficult to calculate. A concrete example is an optimization over the unit ball in a matrix space equipped with operator norm. Our idea consists of a back-projection using quasi-projectors different from the Euclidean best-approximation. In the matrix example, there is another canonical way to force the iterative sequence to stay in the constraint set:
Whenever a point leaves the unit ball, it is divided by its norm. For a given target function, this procedure might introduce spurious stationary points on the boundary. We show that this problem can be circumvented by using a gradient that is tailored to the quasi-projector used for back-projection. We state a general technical compatibility condition between a quasi-projector and a metric used for gradient ascent, prove convergence of stochastic iterative sequences and provide an appropriate metric for the unit-ball example.
Thirdly, a class of learning problems in the sensorimotor loop is defined and motivated. This class of problems is more general than the usual expected reward maximization and is illustrated by numerous examples (like expected reward maximization, maximization of the predictive information, maximization of the entropy and minimization of the variance of a given reward function). We also provide stationarity conditions together with appropriate gradient formulas.
Last but not least, we prove convergence of a stochastic optimization algorithm (as considered in the second topic) applied to a general learning problem (as considered in the third topic). It is shown that the learning algorithm converges to the set of stationary points. Among others, the proof covers the convergence of an improved version of an algorithm for the maximization of the predictive information as proposed by N. Ay, R. Der and K. Zahedi. We also investigate an application to a linear Gaussian dynamic, where the policies are encoded by the unit-ball in a space of matrices equipped with operator norm.
3
Yang, Karren Dai. "Learning causal graphs under interventions and applications to single-cell biological data analysis." Thesis, Massachusetts Institute of Technology, 2021. https://hdl.handle.net/1721.1/130806.
Full textAbstract:
Thesis: S.M., Massachusetts Institute of Technology, Department of Biological Engineering, February, 2021Thesis: S.M., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, February, 2021
Cataloged from the official PDF version of thesis.
Includes bibliographical references (pages 49-51).
This thesis studies the problem of learning causal directed acyclic graphs (DAGs) in the setting where both observational and interventional data is available. This setting is common in biology, where gene regulatory networks can be intervened on using chemical reagents or gene deletions. The identifiability of causal DAGs under perfect interventions, which eliminate dependencies between targeted variables and their direct causes, has previously been studied. This thesis first extends these identifiability results to general interventions, which may modify the dependencies between targeted variables and their causes without eliminating them, by defining and characterizing the interventional Markov equivalence class that can be identified from general interventions. Subsequently, this thesis proposes the first provably consistent algorithm for learning DAGs in this setting. Finally, this algorithm as well as related work is applied to analyze biological datasets.
by Karren Dai Yang.
S.M.
S.M.
S.M. Massachusetts Institute of Technology, Department of Biological Engineering
S.M. Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science
4
Giasemidis, Georgios. "Spectral dimension in graph models of causal quantum gravity." Thesis, University of Oxford, 2013. http://ora.ox.ac.uk/objects/uuid:d0aaa6f2-dd0b-4ea9-81c1-7c9e81a7229e.
Full textAbstract:
The phenomenon of scale dependent spectral dimension has attracted special interest in the quantum gravity community over the last eight years. It was first observed in computer simulations of the causal dynamical triangulation (CDT) approach to quantum gravity and refers to the reduction of the spectral dimension from 4 at classical scales to 2 at short distances. Thereafter several authors confirmed a similar result from different approaches to quantum gravity. Despite the contribution from different approaches, no analytical model was proposed to explain the numerical results as the continuum limit of CDT. In this thesis we introduce graph ensembles as toy models of CDT and show that both the continuum limit and a scale dependent spectral dimension can be defined rigorously. First we focus on a simple graph ensemble, the random comb. It does not have any dynamics from the gravity point of view, but serves as an instructive toy model to introduce the characteristic scale of the graph, study the continuum limit and define the scale dependent spectral dimension. Having defined the continuum limit, we study the reduction of the spectral dimension on more realistic toy models, the multigraph ensembles, which serve as a radial approximation of CDT. We focus on the (recurrent) multigraph approximation of the two-dimensional CDT whose ensemble measure is analytically controlled. The latter comes from the critical Galton-Watson process conditioned on non-extinction. Next we turn our attention to transient multigraph ensembles, corresponding to higher-dimensional CDT. Firstly we study their fractal properties and secondly calculate the scale dependent spectral dimension and compare it to computer simulations. We comment further on the relation between Horava-Lifshitz gravity, asymptotic safety, multifractional spacetimes and CDT-like models.
5
CALIGARIS, SILVIA. "A Causal Graphs - based approach for assessing gender disparities: an application to child health & nutrition in China." Doctoral thesis, Università degli Studi di Milano-Bicocca, 2014. http://hdl.handle.net/10281/83241.
Full textAbstract:
Most of gender statistical measures proposed in the last decades are in fact composite indicators,
i.e. weighted linear combinations of basic statistics such as ratios, percentages etc. Composite indicators then
involves several arbitrary choices -for instance the weighting/aggregating system, variables selection,
standardization affecting both indexes transparency and interpretation. Furthermore gender inequality is a
complex latent phenomenon, a collection of disparate and inter-linked issues that can be hardly caught in a
single indicator. The development of statistical tools and ad hoc models is then required.
The aim of this work is to explore the potential of graphical models as a language able to clearly represent the
complex relationships among variables involved in the statistical measuring the gender disparities. In particular
we will focus on causal graphs allowing to deep and interpret the causal mechanism that may originate gender
gaps as well as to explore the effects of gender tailored policies. Causal models indeed provide transparent
mathematical tools to formulate the assumptions underlying all causal inference, to translate them in term of
joint distribution and to read off the conditional independences using the d-separation criterion (Pearl 2000). It
is thus possible deriving causal effects in non-experimental studies, representing policies’ effects and
interventions through the do operator, controlling confounders and interpreting counterfactuals.
We show the potential of such models through an application to real data from China Health and Nutrition
Survey 2011 ; in particular we explore the eventual existence of gender discrimination in children’ nutrition and
health as possible indicator of preference for sons. The analysis takes in exam socio-demographic, economical
as well as biological variables. Resorting to the PC algorithm and the IDA algorithm, we aim to learn the
underlying causal structure and to estimate causal effect of siblings on children’ nutrition from observational
data.
6
Bernigau, Holger [Verfasser], Nihat [Akademischer Betreuer] Ay, Nihat [Gutachter] Ay, and Shun-ichi [Gutachter] Amari. "Causal Models over Infinite Graphs and their Application to the Sensorimotor Loop : Causal Models over Infinite Graphs and their Application to theSensorimotor Loop / Holger Bernigau ; Gutachter: Nihat Ay, Shun-ichi Amari ; Betreuer: Nihat Ay." Leipzig : Universitätsbibliothek Leipzig, 2015. http://d-nb.info/1239565127/34.
Full text
7
Chong, Hogun. "A causal model of linkages among strategy, structure, and performance using directed acyclic graphs: A manufacturing subset of Fortune 500 industrials 1990-1998." Texas A&M University, 2003. http://hdl.handle.net/1969.1/58.
Full textAbstract:
This research explored the causal relationships among strategies, corporate structure, and performance of the largest U.S. non-financial firms using Directed Acyclic Graphs (DAGs). Corporate strategies and structure have been analyzed as major variables to influence corporate performance in management and organizational studies. However, their causal relationships in terms of which variables are leaders and followers, as well as the choices of variables to configure them, are controversial. Finding of causal relationships among strategic variables, structural variables, and corporate performance is beneficial to researchers as well as corporate mangers. It provides guidance to researchers how to build a model in order to measure influences from one variable to the other, lowering the risk of drawing spurious conclusions. It also provides managers a prospect of how certain important variables would change by making a certain strategic decision. Literatures from agency theory, transactional cost economics, and traditional strategic management perspective are used to suggest variables essential to analyze corporate performance. This study includes size and multi-organizational ownership hierarchy as variables to configure corporate structure. The variables to configure corporate strategies are unrelated and related diversification, ownership by institutional investors, debt, investment in R&D, and investment in advertisement.
The study finds that most of the variables classified as corporate strategy and corporate structure variables are either direct or indirect causes of corporate accounting performance. Generally, results supports the relational model: corporate structure® corporate strategy® corporate performance. Ownership hierarchy structure, unrelated diversification, advertising expenses, and R&D intensity have direct causal influences on corporate accounting performance. Size and related diversification affected corporate accounting performance indirectly, both through ownership hierarchy structure. Theoretical causal relationships from agency theory are less supported than those from transaction cost economics and traditional strategic management perspective. Further my study suggests that, in general, good corporate performance in 1990s was mainly achieved by internal expansion through investment in R&D and advertisement, rather than external expansion of firms through unrelated diversification, related diversification, and expansion of ownership hierarchy.
8
Aka, Niels Mariano [Verfasser]. "Three Essays on Model Selection in Time Series Econometrics : Model Averaging, Causal Graphs, and Structural Identification / Niels Mariano Aka." Berlin : Freie Universität Berlin, 2021. http://d-nb.info/1229436685/34.
Full text
9
Martiel, Simon. "Approches informatique et mathématique des dynamiques causales de graphes." Thesis, Nice, 2015. http://www.theses.fr/2015NICE4043/document.
Full textAbstract:
Le modèle des automates cellulaires constitue un des modèles le mieux établi de physique discrète sur espace euclidien. Ils implantent trois symétries fondamentales de la physique: la causalité, l'homogénéité et la densité finie de l'information. Bien que l'origine des automates cellulaires provienne de la physique, leur utilisation est très répandue comme modèles de calcul distribué dans l'espace (machines auto-réplicantes, problèmes de synchronisation,...), ou bien comme modèles de systèmes multi-agents (congestion du trafic routier, études démographiques,...). Bien qu'ils soient parmi les modèles de calcul distribué les plus étudiés, la rigidité de leur structure interdit toute extension triviale vers un modèle de topologie variant dans le temps, qui se trouve être un prérequis fondamental à la modélisation de certains phénomènes biologiques, sociaux ou physiques, comme par exemple la discrétisation de la relativité générale. Les dynamiques causales de graphes généralisent les automates cellulaires aux graphes arbitraires de degré borné et pouvant varier dans le temps. Dans cette thèse, nous nous attacherons à généraliser certains des résultats fondamentaux de la théorie des automates cellulaires. En munissant nos graphes d'une métrique compacte, nous présenterons deux approches différentes du modèle. Une première approche axiomatique basée sur les notions de continuité et d'invariance par translation, et une deuxième approche constructive, où une règle locale est appliquée en parallèle et de manière synchrone sur l'ensemble des sommets du grapheCellular Automata constitute one of the most established model of discrete physical transformations that accounts for euclidean space. They implement three fundamental symmetries of physics: causality, homogeneity and finite density of information. Even though their origins lies in physics, they are widely used to model spatially distributed computation (self-replicating machines, synchronization problems,...), as well as a great variety of multi-agents phenomena (traffic jams, demographics,...). While being one of the most studied model of distributed computation, their rigidity forbids any trivial extension toward time-varying topology, which is a fundamental requirement when it comes to modelling phenomena in biology, sociology or physics: for instance when looking for a discrete formulation of general relativity. Causal graph dynamics generalize cellular automata to arbitrary, bounded degree, time-varying graphs. In this work, we generalize the fundamental structure results of cellular automata for this type of transformations. We endow our graphs with a compact metric space structure, and follow two approaches. An axiomatic approach based on the notions of continuity and shift-invariance, and a constructive approach, where a local rule is applied synchronously on every vertex of the graph. Compactness allows us to show the equivalence of these two definitions, extending the famous result of Curtis-Hedlund-Lyndon’s theorem. Another physics-inspired symmetry is then added to the model, namely reversibility
10
Encardes, Nicole A. "Causal factors of Macrophoma rot observed on Petit Manseng grapes." Thesis, Virginia Tech, 2020. http://hdl.handle.net/10919/99083.
Full textAbstract:
Macrophoma rot is a general term for fruit rots of Vitis spp. caused by the fungus Neofusicoccum ribis (syn. Botryosphaeria ribis) or closely related or renamed taxa, including Botryosphaeria dothidea. While mainly observed as a fruit pathogen of muscadine grape, the disease has recently been observed on bunch grapes in Virginia. Isolates (N = 835) were collected from Petit Manseng fruit clusters from seven Virginia vineyards in 2018 and 2019. A subset of these isolates was sequenced using three primer sets (ITS, RPB2, and EF). The preliminary result showed a single taxonomic strain of N. ribis. A controlled inoculation study of Petit Manseng clusters verified that infection could occur anytime between bloom and 2 weeks post-veraison; however, both the mean cluster incidence and the severity of Macrophoma rot did not differ from each other at any growth stage during the season. A season-long cluster exposure experiment showed that any amount of sun exposure significantly increased Macrophoma rot severity compared to shaded clusters, and that full sun exposure was associated with greatest rot severity. This finding contravenes current management recommendations for Macrophoma rot, and it raises yet unanswered questions as to why exposed clusters are more susceptible to Macrophoma rot than are shaded clusters. An in vitro fungicide assay study using nine fungicides identified captan, thiophanate-methyl, and tetraconazole as potential candidates for management of Macrophoma rot which need to be investigated further.Master of Science in Life Sciences
Macrophoma rot is a general term for fruit rots of grapes caused by the pathogenic fungi in the family Botryosphaeriaceae. The rot is mainly observed on Muscadine grapes, but recently more cases were found on a wine grape cultivar Petit Manseng in Virginia. Macrophoma rot symptoms begin as dark brown, circular lesions on the surface of the berry and look similar to sunburn and other fruit rots. As the disease progresses, the lesion envelopes the entire berry and black fruiting bodies develop. Severe cases may lead to crop loss. The same group of pathogens is also associated with rots on other crops including apple, pear, olive, and kiwis. Very little is known about the disease cycle and the control of Macrophoma rot, therefore, an investigation into this fungal pathogen was needed. Multiple studies with the wine grape variety Petit Manseng were conducted during the 2018-2019 growing seasons, including a survey, leaf removal trial, and an inoculation study. Results showed that a species called Neofusicoccum ribis was found in vineyards across northern and central Virginia based on the genetic identification of fungal isolates collected at seven vineyards in those areas. Macrophoma symptoms were observed to be more prevalent and severe in more exposed clusters based on a leaf removal experiment. An artificial inoculation experiment revealed that grape clusters are susceptible to Neofusicoccum ribis at any time during the season. Based on the screening of nine fungicides, three chemicals (captan, thiophanate-methyl, and tetraconazole) showed promising results as possible management tools for Macrophoma rot. The knowledge collected will lead to an increase in understanding of this fungal pathogen and to further studies to manage Macrophoma rot.
11
Munch, Mélanie. "Améliorer le raisonnement dans l'incertain en combinant les modèles relationnels probabilistes et la connaissance experte." Thesis, université Paris-Saclay, 2020. http://www.theses.fr/2020UPASB011.
Full textAbstract:
Cette thèse se concentre sur l'intégration des connaissances d'experts pour améliorer le raisonnement dans l'incertitude. Notre objectif est de guider l'apprentissage des relations probabilistes avec les connaissances d'experts pour des domaines décrits par les ontologies.Pour ce faire, nous proposons de coupler des bases de connaissances (BC) et une extension orientée objet des réseaux bayésiens, les modèles relationnels probabilistes (PRM). Notre objectif est de compléter l'apprentissage statistique par des connaissances expertes afin d'apprendre un modèle aussi proche que possible de la réalité et de l'analyser quantitativement (avec des relations probabilistes) et qualitativement (avec la découverte causale). Nous avons développé trois algorithmes à travers trois approches distinctes, dont les principales différences résident dans leur automatisation et l'intégration (ou non) de la supervision d'experts humains.L'originalité de notre travail est la combinaison de deux philosophies opposées : alors que l'approche bayésienne privilégie l'analyse statistique des données fournies pour raisonner avec, l'approche ontologique est basée sur la modélisation de la connaissance experte pour représenter un domaine. La combinaison de la force des deux permet d'améliorer à la fois le raisonnement dans l'incertitude et la connaissance experteThis thesis focuses on integrating expert knowledge to enhance reasoning under uncertainty. Our goal is to guide the probabilistic relations’ learning with expert knowledge for domains described by ontologies.To do so we propose to couple knowledge bases (KBs) and an oriented-object extension of Bayesian networks, the probabilistic relational models (PRMs). Our aim is to complement the statistical learning with expert knowledge in order to learn a model as close as possible to the reality and analyze it quantitatively (with probabilistic relations) and qualitatively (with causal discovery). We developped three algorithms throught three distinct approaches, whose main differences lie in their automatisation and the integration (or not) of human expert supervision.The originality of our work is the combination of two broadly opposed philosophies: while the Bayesian approach favors the statistical analysis of the given data in order to reason with it, the ontological approach is based on the modelization of expert knowledge to represent a domain. Combining the strenght of the two allows to improve both the reasoning under uncertainty and the expert knowledge
12
Cortes, Taísa Rodrigues. "Utilização de diagramas causais em confundimento e viés de seleção." Universidade do Estado do Rio de Janeiro, 2014. http://www.bdtd.uerj.br/tde_busca/arquivo.php?codArquivo=8442.
Full textAbstract:
Conselho Nacional de Desenvolvimento Científico e TecnológicoApesar do crescente reconhecimento do potencial dos diagramas causais por epidemiologistas, essa técnica ainda é pouco utilizada na investigação epidemiológica. Uma das possíveis razões é que muitos temas de investigação exigem modelos causais complexos. Neste trabalho, a relação entre estresse ocupacional e obesidade é utilizada como um exemplo de aplicação de diagramas causais em questões relacionadas a confundimento. São apresentadas etapas da utilização dos diagramas causais, incluindo a construção do gráfico acíclico direcionado, seleção de variáveis para ajuste estatístico e a derivação das implicações estatísticas de um diagrama causal. A principal vantagem dos diagramas causais é tornar explícitas as hipóteses adjacentes ao modelo considerado, permitindo que suas implicações possam ser analisadas criticamente, facilitando, desta forma, a identificação de possíveis fontes de viés e incerteza nos resultados de um estudo epidemiológico.
Despite the increasing recognition of the potential of causal diagrams by epidemiologists, this technique has not been widely used in epidemiological research. One possible reason is that many research topics require complex causal models. In this article, the relationship between occupational stress and obesity is used as an example of application of causal diagrams on confounding. Some steps are presented, including the construction of the directed acyclic graph, the selection of variables for statistical control and the derivation of the statistical implications of a causal diagram. The main advantage of causal diagrams is to make the assumptions explicit, thus facilitating critical evaluations and the identification of possible sources of bias and uncertainty in the results of an epidemiological study.
13
Lafon, Marc. "Modélisation de la propagation de l'information cérébrale par graphes causaux qualitatifs." Phd thesis, Université Paul Sabatier - Toulouse III, 2000. http://tel.archives-ouvertes.fr/tel-00010176.
Full textAbstract:
Des mesures de l'activité du cerveau humain peuvent être obtenues par diverses techniques d'imagerie neuro-fonctionnelle. Dans le domaine des neurosciences, les études menées grâce à ces nouvelles techniques sont appelées études d'activation. Elles mettent en évidence l'organisation en réseau des aires cérébrales mises en jeux lors de la réalisation de tâches cognitives particulières. La complexité du traitement de l'information cérébrale et l'incomplétude de l'observation rendent difficile l'interprétation de ce fonctionnement et rendent nécessaire la construction d'un outil de modélisation et de simulation. La nécessité de comprendre l'organisation fonctionnelle du cerveau humain, nous impose en plus de recourir à des formalismes permettant d'en avoir une représentation explicite. Dans cette optique, nous proposons un formalisme original, basé sur des graphes d'influences causales et une représentation qualitative de l'information. Les graphes d'influences sont tout à fait appropriés à la représentation de la double organisation du cerveau, en réseaux anatomiques (structurels) et fonctionnels. L'originalité de notre approche est double. D'une part, nous proposons une architecture hiérarchique fondée sur la notion de composant et qui permet de réduire la complexité des modèles tout en augmentant leur réutilisabilité. D'autre part, nous utilisons un formalisme unique pour décrire les niveaux structurels et fonctionnels cérébraux. La modélisation explicite de l'information cérébrale nous a conduit à proposer une représentation qualitative mixte, sous la forme de couples (intervalle, symbole). Ceci permet à la fois de prendre en compte l'amplitude de cette information, bruitée par les techniques d'imagerie, et sa catégorie. Nous proposons le simulateur BIOCAEN, fondé sur le formalisme ci-dessus. La simulation consiste en la propagation des influences à chaque instant d'une horloge discrète. Ce mécanisme permet de prendre en compte les processus automatiques qui constituent la majeure partie du fonctionnement cérébral. La contribution de notre travail à l'exploration du fonctionnement du cerveau humain réside dans la mise à disposition des chercheurs en neurosciences, d'un formalisme permettant d'exprimer explicitement, à un haut niveau d'abstraction, les hypothèses sur la propagation de l'information cérébrale et d'un simulateur permettant de les mettre en œuvre. Les hypothèses sont décrites dans des modèles où l'architecture causale offre un support à l'expression des réseaux anatomiques d'aires cérébrales, chacune d'entre elles étant représentée par un sous-réseau fonctionnel de processeurs d'information. La validation des hypothèses est réalisée par la confrontation des résultats de la simulation du modèle par BIOCAEN aux données de l'imagerie neuro-fonctionnelle.
14
Macedo, Patrícia Alexandra Pires. "Models and tools for value systems analysis in collaborative environments." Doctoral thesis, Faculdade de Ciências e Tecnologia, 2011. http://hdl.handle.net/10362/6106.
Full textAbstract:
Dissertation to obtain the degree of Doctor in Electrical and Computer Engineering, specialization of Collaborative NetworksParticipation in collaborative networks is vital for small and medium-sized enterprises to survive in the current market, bringing them several benefits. However,participation in collaborative networks also involves risks and often consortia fail due to internal conflicts. Conflicts can be originated by different prioritization of values and different perceptions of outcomes. The perception of outcomes is, to some extent, subjective given that it depends on the preferences of the subject and how exchanges are evaluated. Therefore, the establishment of a common Value System or the effort to align the Value Systems of network members can play an important role in the collaboration sustainability. Although the topic of values and values alignment has been studied within the scope of various scientific disciplines, there is still no common understanding on these concepts and the literature does not include any suitable models to formally represent and analyze Value Systems within the scope of collaborative networks. This thesis proposes a set of models and formal mechanisms for specifying and analyzing Value Systems in collaborative networks. The development of models and methods followed a hybrid approach, where qualitative and quantitative techniques are used in order to represent and analyze the Value System. A web application was designed and a prototype developed in order to show that the models and methods proposed can be implemented by a computer program and can be integrated into a single framework in order to support Value Systems management within the scope of collaborative networks. The application of a multifaceted and systematic validation strategy, supported by the “Square Validation Framework” brought together a set of preliminary results that attest the theoretical and practical relevance of the proposed approach and allow us to conclude that: (i) it is possible to define and analyze Value Systems in collaborative networks, considering the economic and sociologic approach, in an integrated and unambiguous way, (ii) the potential impacts between Value Systems in collaborative environments can be inferred if the typical influences among core values are known and the preferences of the actors, regarding those values, are identified; (iii) the identification and assessment of Value Systems misalignments would be improved if qualitative and quantitative assessment methods integrating the notion of shared-values, potential for conflict and positive impacts were developed.
15
Asvatourian, Vahé. "Apports de la modélisation causale dans l’évaluation des immunothérapies à partir de données observationnelles." Thesis, Université Paris-Saclay (ComUE), 2018. http://www.theses.fr/2018SACLS427/document.
Full textAbstract:
De nouveaux traitements comme l’immunothérapie ont été proposés en oncologie. Ils sont basés sur les mécanismes de régulation du système immunitaire. Cependant tous les patients ne répondent pas à ces nouveaux traitements. Afin de pouvoir les identifier, on mesure l’association des marqueurs immunologiques exprimés à la réponse au traitement ainsi qu’à la toxicité à l’instaurationdu traitement et leur évolution sous traitement. En situation observationnelle, l’absence de tirage au sort empêche la comparabilité des groupes et l'effet mesuré est juste une mesure d'association. Les méthodes d’inférence causalepermettent dans certains cas, après avoir identifié les sources de biais de par la construction de diagrammes acycliques dirigés (DAG), d'atteindre l’interchangeabilité conditionnelle entre exposés et non exposés etpermettent l’estimation d’effets causaux. Dans les cas les plus simples où le nombre de variables est faible, il est possible de dessiner leDAG à partir d’expertise. Dans les situations où le nombre de variables explosent, des algorithmes d’apprentissage ont été proposés pour retrouver la structure de ces graphes. Néanmoins ces algorithmes font d’une part l’hypothèse qu’aucune information n’est connue et n’ont été développés que dans les cas où les covariables sont mesurés à un seul temps. L’objectif de cette thèse est donc de développer ces méthodes d’apprentissages de graphes à des données répétées, puis d’intégrer des connaissances a priori pour améliorer l’estimation de ceux-ci. Une fois les graphes appris les modèles causaux peuvent être appliqués sur les biomarkers immunologiques répétés pour détecter ceux qui sont associés à laréponse et/ou la toxicitéIn oncology, new treatments such as immunotherapy have been proposed, which are based on regulation of the immune system. However, not all treated patient have a long-term benefit of the treatment. To identify those patients who benefit most, we measured markers of the immune system expressed at treatment initiation and across time. In an observational study, the lack of randomization makes the groups not comparable and the effect measured is just an association. In this context, causal inference methods allow in some cases, after having identified all biases by constructing a directed acyclic graph (DAG), to get close to the case of conditional exchangeability between exposed and non-exposed subjects and thus estimating causal effects.In the most simple cases, where the number of variables is low, it is possible to draw the DAG with experts’ beliefs. Whereas in the situation where the number of variables rises, learning algorithms have been proposed in order to estimate the structure of the graphs. Nevertheless, these algorithms make the assumptions that any a priori information between the markers is known and have mainly been developed in the setting in which covariates are measured only once. The objective of this thesis is to develop learning methods of graphs for taking repeated measures into account, and reduce the space search by using a priori expert knowledge. Based on these graphs, we estimate causal effects of the repeated immune markers on treatment response and/or toxicity
16
Giraud, Frédéric. "Modélisation causale et commande d'un actionneur piézo-électrique à onde progressive." Lille 1, 2002. https://pepite-depot.univ-lille.fr/LIBRE/Th_Num/2002/50376-2002-109-110.pdf.
Full textAbstract:
Les moteurs piézo-électriques à onde progressive, grâce à leur caractéristique fort couple - basse vitesse et leur faible encombrement, proposent une alternative intéressante aux machines électromagnétiques dans les systèmes automatisés. Cependant, leur structure vibrante et les caractéristiques tribologiques complexes à l'interface stator-rotor les rendent difficiles à commander. C'est afin de déterminer les lois de commande adaptées que nous avons d'abord modélisé de manière causale le moteur. Nous avons discuté de la pertinence de ce modèle, soit par comparaison avec un modèle plus complexe et plus précis, soit expérimentalement par des relevés sur un moteur commercialisé. Cette modélisation a conduit d'une part à l'élaboration d'un protocole d'identification des paramètres du moteur insensible aux non linéarités inhérentes aux matériaux piézo-électriques. D'autre part, le Graphe Informationnel Causal, utilisé pour la représentation du modèle, a permis, par son inversion, l'établissement des lois de commande. Nous avons alors été capables d'établir plusieurs stratégies d'autopilotage originales; nous n'en avons retenu qu'une, qui a été mise en œuvre sur un banc expérimental. Cette commande a été testée sur un dispositif à retour d'effort, car cette application à basse vitesse fait partie du domaine privilégié de l'utilisation des moteurs piézo-électriques.
17
Giraud, Frédéric Lemaire-Semail Betty. "Modélisation causale et commande d'un actionneur piézo-électrique à onde progressive." [S.l.] : [s.n.], 2002. http://www.univ-lille1.fr/bustl-grisemine/pdf/extheses/50376-2002-109-110.pdf.
Full text
18
LAFON, MARC. "Modelisation de la propagation de l'information cerebrale par graphes causaux qualitatifs." Toulouse 3, 2000. http://www.theses.fr/2000TOU30038.
Full textAbstract:
Dans les etudes d'activation, l'activite du cerveau humain est mesuree grace a diverses techniques d'imagerie neuro-fonctionnelle. Ces etudes mettent en evidence les reseaux d'aires cerebrales mis en jeux lors de la realisation de taches cognitives. Mais la complexite du traitement cerebral et l'incompletude de l'observation compliquent l'interpretation de ce fonctionnement et necessitent la construction d'un outil de modelisation et de simulation. Pour parvenir a une comprehension de l'organisation fonctionnelle du cerveau humain un recours a une representation explicite est necessaire. Dans cette optique, nous proposons un formalisme original, fonde sur des graphes d'influences causales et une representation qualitative de l'information. Les graphes nous permettent de representer la double organisation du cerveau, en reseaux anatomiques (structurels) et fonctionnels. Nous proposons une architecture hierarchique qui reduit la complexite des modeles tout en augmentant leur reutilisabilite et nous utilisons un formalisme homogene pour decrire les deux niveaux. Une modelisation explicite de l'information cerebrale nous conduit a une representation qualitative mixte, sous la forme d'un couple (intervalle, symbole) pour prendre en compte l'amplitude et la categorie de l'information. Fondes sur ce formalisme, nous proposons le simulateur biocaen limite aux processus automatiques qui constituent la majeure partie du fonctionnement cerebral. La contribution de notre travail reside dans la mise a disposition d'un environnement permettant d'exprimer, a un haut niveau d'abstraction, des hypotheses sur la propagation de l'information cerebrale. Elles sont decrites par des modeles ou l'architecture causale sert de support a l'expression des reseaux anatomiques d'aires cerebrales, chacune etant representee par un sous-reseau fonctionnel. La confrontation des resultats de la simulation aux donnees de l'imagerie neuro-fonctionnelle permet la validation des hypotheses.
19
Xia, Xiaoguang. "Contribution à l'analyse du comportement dynamique et énergétique d'un système modélisé par un bond graph bi causal." Lyon, INSA, 2000. http://www.theses.fr/2000ISAL0114.
Full textAbstract:
Les travaux présentés dans cette thèse concernent l'analyse du comportement dynamique et énergétique d'un système linéaire et stationnaire modélisé par un bond graph bicausal. Notre contribution porte sur l'introduction du bond graph du système adjoint. Et du système dual d'un système dont le bond graph causal est donné. Les notions de bond graph bicausal inverse et direct sont utilisées pour développer une séries de procédures relatives au calcul : de la fonction de transfert entre deux variables internes du système, de la matrice d'opérateurs différentiels de transmission, de la matrice de répartition de l'énergie en régime sinusoïdal établi et la relation entre les puissances complexes associées à deux liens de puissance quelconques du bond graph. La justification de certaines procédures développées repose sur le concept original de matrice bond graph. •L'étude de la répartition de l'énergie en régime sinusoïdal établi est basée sur la notion classique de puissance complexe et a donné lieu à l'utilisation du bond graph complexe associé à un bond graph donné. Chaque procédure proposée est illustrée par un exemple. Des applications au dimensionnement d'un système dans le domaine fréquentiel sont envisagéesThe work presented in this thesis deal with the analysis of the dynarnic and energy performance of a linear and stationary system modeled by a bicausal bond graph. Our contribution lies in the introduction of the bond graph of the adjoin system and dual system associated with a system whose bond graph is given. The notions of inverse and direct bicausal bond graphs are used for developing a series of procedures such as: calculation of the transfer function between two internal variables of the system, determination of the transmission matrix with differential operator, formulation of energy distribution matrix in the mode of sinusoidal established and establishment of the relation between the complex powers associated with two power bonds. The demonstration of certain procedures developed rest on the original conception of bond graph matrix. The study of energy distribution in the mode of sinusoidal established is based on the classical notion of complex power and is reduced with the help of the introduction of the complex bond graph associated with a given bond graph. Each procedure proposed is illustrated by an example. The applications in sizing of a system in the frequency domain is to be considered
20
Freund, Alexander J. "The Necessity and Challenges of Automatic Causal Map Processing: A Network Science Perspective." Miami University / OhioLINK, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=miami1619545359648916.
Full text
21
Xu, Ning. "Accurate variable selection and causal structure recovery in high-dimensional data." Thesis, University of Sydney, 2020. https://hdl.handle.net/2123/22920.
Full textAbstract:
From the perspective of econometrics, an accurate variable selection method greatly enhances the reliability of causal analysis and interpretation of the estimators, espe- cially in a world of ever-expanding data dimensions. While variable selection methods in machine learning and statistics have been developed rapidly and applied widely in different branches of data science in the last decade, they have been more slowly adopted in econometrics. Nevertheless, the machine learning methods, including lasso, forward regression, cross-validation and marginal correlation ranking (also called vari- able screening) are subject to a range of issues that may result in errors in variable selection and inaccurate causal interpretation. I propose two new variable-selection methods that significantly mitigate the issues with existing techniques and that provide accurate variable selection and reliable causal structure estimation in high-dimensional data. In Chapter 1, I develop bounds for cross-validation errors that may be used as a criterion for variable selection with many existing learning algorithms (including lasso, forward regression and variable screen- ing), yielding a sparse and stable model that retains all of the relevant variables. In Chapter 2, I develop an entirely new learning algorithm for variable selection— subsample-ordered least-angle regression (solar)—and show in simulations that solar out-performs coordinate descent and lars-lasso in terms of the sparsity, stability, ac- curacy, and robustness of variable selection. In Chapter 3 I demonstrate the superior variable-selection performance of solar using real-world data from two completely dif- ferent samples: prostate cancer patients and house prices. I also show that combining solar variable selection with linear probabilistic graph learning yields a plausible, data- driven method to recover causal structure in data.
22
Gomand, Julien. "Analyse de systemes multi-actionneurs paralleles par une approche graphique causale : application a un processus electromecanique de positionnement rapide." Paris, ENSAM, 2008. http://www.theses.fr/2008ENAM0043.
Full textAbstract:
L'augmentation des cadences de production nécessite la recherche de performances dynamiques toujours plus élevées pour les dispositifs de positionnement, conduisant à l'utilisation d'axes multi-actionneurs parallèles. Une configuration bi-actionnée en gantry est une solution mécanique qui permet de conférer une accélération élevée à l'axe, mais pose la problématique de la synchronisation des actionneurs mis en collaboration. Un état de l'art de la commande de tels dispositifs électromécaniques met en évidence un manque général de prise en compte du couplage mécanique entre les deux actionneurs, conduisant à des architectures de commande complexes sur le plan structurel et du réglage. L'approche proposée dans cette étude constitue une alternative à ces commandes en mettant en œuvre une approche graphique structurée selon les lois de la causalité physique. Dans un premier temps, l'axe en gantry sert de support à l'étude de propriétés graphiques du formalisme Graphe Informationnel Causal (GIC) développé au L2EP. L'ordonnancement des représentations selon la causalité intégrale conduit à une simplification globale de l'analyse structurelle des systèmes physiques, facilitant le choix des variables d'état indépendantes à contrôler. La modélisation dynamique proposée pour l'axe considéré, associée à des méthodes d'identification expérimentale, est exploitée pour la déduction d'architectures de commande et de méthodes de réglage permettant d'améliorer les performances de synchronisation dynamique des deux actionneurs par une meilleure gestion du couplage. Les problématiques de la modélisation et de la compensation des ondulations de force générées par les moteurs linéaires synchrones à aimants permanents, ainsi que de la gestion énergétique de l'ensemble de l'axe sont également abordéesAbstract : High-dynamic positioning systems are required to increase production rates. A dual-drive gantry configuration is a mechanical solution which offers higher acceleration levels. However, the two parallel collaborative linear actuators have to be perfectly synchronized. Actual control solutions for such an electromechanical device shows complex control structures, without real consideration of the mechanical coupling, which leads to compensator tuning difficulties. The proposed approach is model based, using physical causality ordering laws. First of all, the gantry axis is used as an example to study the graphical properties of the Causal Ordering Graph (COG) formalism, developed by the L2EP laboratory. It simplifies the structural analysis of physical systems, and facilitates the identification of independent state variables to be controlled. A dynamic model of the considered gantry axis is then established, and experimental identification methods are developed. Inversion-based decoupling motion control allows simple tuning methods and leads to improved synchronizing performances. The compensation of force ripples generated by permanent magnet linear synchronous motors and the global system energetic control are also studied
23
Araujo, João Paulo de. "A communication-efficient causal broadcast publish/subscribe system." Electronic Thesis or Diss., Sorbonne université, 2019. http://www.theses.fr/2019SORUS081.
Full textAbstract:
La Publication/Abonnement (Publish/Subscribe, Pub/Sub) est un paradigme qui permet aux nœuds d'un système distribué de diffuser des informations de manière asynchrone. Cette thèse s'intéresse aux systèmes de Pub/Sub basés sur des sujets (topic-based), en adressant les problèmes de performances et de contention existant dans plusieurs approches reposant sur des arbres. Les solutions proposées utilisent la construction d'arbres couvrants regroupant les abonnés et dont les racines sont les émetteurs. Les arbres associés à différentes sources sont organisés différemment. La première contribution de la thèse propose un protocole de diffusion causal agrégeant des messages et dans lequel aucun temporisateur n'est nécessaire. Le protocole regroupe les messages en un seul message sans utiliser des temporisateurs en tirant parti du délai de livraison supplémentaire imposé à un nœud lorsque les messages sont reçus en dehors de l'ordre causal ainsi que des intersections existantes entre des arbres couvrants. La deuxième contribution est un système de Pub/Sub par sujet, VCube-PS, qui assure l'ordre de traitement causal des messages publiés sur un même sujet et gère efficacement la publication de messages sur des sujets très populaires ("hot topics"). Les résultats des simulations confirment que le protocole d'agrégation causale proposé réduit le trafic réseau ainsi que des latences de livraison, en limitant la contention de messages. Comparé à une approche utilisant un seul arbre par sujet, VCube-PS repartit mieux la charge lors de publications massives sur des "hot topic"The Publish/Subscribe (Pub/Sub) paradigm enables nodes of a distributed system to disseminate information asynchronously. This thesis investigates how to provide a communication-efficient topic-based Pub/Sub system by addressing the problems of traffic overhead and message contention, present in several tree-based solutions. The proposed contributions build distributed spanning trees on top of a hypercube-like topology, such that the source of each message is the root of its own dynamically built spanning tree. Trees rooted at different nodes are differently organized. Initially, it is proposed a causal broadcast protocol which reduces network traffic by aggregating messages without the use of timers. It exploits the causal relation between messages and path intersections between different trees. Different from existing timer-based approaches, it does not increase delivery latency. The second contribution is a topic-based Pub/Sub system, VCube-PS, which ensures causal delivery order for messages published to the same topic and efficiently supports publication of messages to "hot topics'', i.e., topics with high publication rates. Simulation results confirm that the proposed causal aggregation protocol reduces network traffic as well as delivery latencies since there is less message contention. Compared to an approach that uses one single tree per topic, VCube-PS performs better when there is a high publication rate per topic since it provides load balancing of publication
24
Borrel, Vincent. "Une approche intégrée pour la mobilité humaine dans les réseaux auto-organisables : causes et conséquences." Paris 6, 2008. http://www.theses.fr/2008PA066551.
Full textAbstract:
Cette thèse se concentre sur deux aspects de la mobilité dans les réseaux: les modèles synthétiques des déplacements humains, vu sous l'angle de leurs causes sociologiques, et la classification de cette mobilité en fonction de son impact sur les choix de routage des données. La mobilité des individus influence leurs possibilités de communication sans fil, modifiant les conditions et performance des liens de données qui les rattachent aux autres. Dans l'optique de réseaux sans-fil futurs où l'information passe de proche en proche, au gré des rencontres entre humains, la compréhension de cette mobilité joue un rôle crucial. Ainsi, de la connaissance et l'exploration de caractéristiques de cette mobilité, et de l'impact de celle-ci sur le routage et le transport collectif des données en général, pourront découler de nombreuses améliorations à ce problème délicat: s'adapter aux incessants mouvements des hommes. Cette thèse explore deux facettes de ce vaste problème, qui sont les causes sociales de la mobilité, et les conséquences de celle-ci sur le type de routage accessible au réseau évolutif en résultant. L'aspect causal de la mobilité, sous sa forme sociale fait l'objet de la première partie ou deux travaux sont présentés. La seconde partie de ce manuscript traite des conséquences de la mobilité. Elle est l'objet d'un troisième travail, qui est une taxonomie des possibilités de communication offertes par un réseau, associée a une méthode de classification de l'impact de la mobilité sur ces possibilités.
25
Kalainathan, Diviyan. "Generative Neural Networks to infer Causal Mechanisms : algorithms and applications." Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLS516.
Full textAbstract:
La découverte de relations causales est primordiale pour la planification, le raisonnement et la décision basée sur des données d'observations ; confondre corrélation et causalité ici peut mener à des conséquences indésirables. La référence pour la découverte de relations causales est d'effectuer des expériences contrôlées. Mais dans la majorité des cas, ces expériences sont coûteuses, immorales ou même impossible à réaliser. Dans ces cas, il est nécessaire d'effectuer la découverte causale seulement sur des données d'observations. Dans ce contexte de causalité observationnelle, retrouver des relations causales introduit traditionellement des hypothèses considérables sur les données et sur le modèle causal sous-jacent. Cette thèse vise à relaxer certaines de ces hypothèses en exploitant à la fois la modularité et l'expressivité des réseaux de neurones pour la causalité, en exploitant à la fois et indépendences conditionnelles et la simplicité des méchanismes causaux, à travers deux algorithmes. Des expériences extensives sur des données simulées et sur des données réelles ainsi qu'une analyse théorique approfondie prouvent la cohérence et bonne performance des approches proposéesCausal discovery is of utmost importance for agents who must plan, reason and decide based on observations; where mistaking correlation with causation might lead to unwanted consequences. The gold standard to discover causal relations is to perform experiments.However, experiments are in many cases expensive, unethical, or impossible to realize. In these situations, there is a need for observational causal discovery, that is, the estimation of causal relations from observations alone.Causal discovery in the observational data setting traditionally involves making significant assumptions on the data and on the underlying causal model.This thesis aims to alleviate some of the assumptions made on the causal models by exploiting the modularity and expressiveness of neural networks for causal discovery, leveraging both conditional independences and simplicity of the causal mechanisms through two algorithms.Extensive experiments on both simulated and real-world data and a throughout theoretical anaylsis prove the good performance and the soundness of the proposed approaches
26
Pigache, François Lemaire-Semail Betty. "Modélisation causale en vue de la commande d'un translateur piézoélectrique plan pour une application haptique." Villeneuve d'Ascq : Université des sciences et technologies de Lille, 2007. https://iris.univ-lille1.fr/dspace/handle/1908/398.
Full textAbstract:
Reproduction de : Thèse de doctorat : Génie électrique : Lille 1 : 2005.N° d'ordre (Lille 1) : 3612. Résumé en français et en anglais. Titre provenant de la page de titre du document numérisé. Bibliogr. p. 161-167.
27
Malaguti, Giulia. "Analyses théoriques de l'expansion des familles de gènes impliqués dans des maladies dominantes." Thesis, Paris 6, 2014. http://www.theses.fr/2014PA066319/document.
Full textAbstract:
Les familles de gènes impliqués dans le cancer et autres maladies génétiques se sont beaucoup élargies via deux Duplications Globales de Génome (DGG) qui ont eu lieu à l'origine des vertébrés. La rétention des copies de ces gènes implique une susceptibilité plus grande aux maladies génétiques et constitue une énigme du point de vue de l'évolution. Dans cette thèse, nous avons généralisé des modèles classiques de génétique des populations pour révéler le mécanisme non-adaptatif qui a conduit à cette conservation de gènes potentiellement délétères chez les vertébrés. Nous avons résolu un modèle déterministe haploïde, nous avons étendu ce modèle à des génomes diploïdes et nous avons analysé les effets de taille finie des populations et de la sélection positive par une approche stochastique. Les résultats montrent, en accord avec les données génomiques du cancer chez l'homme, que les copies DGG susceptibles aux mutations délétères dominantes sont conservées indirectement via la sélection de purification dans les espèces post-DGG, qui présentent nécessairement une incompatibilité de ploïdie avec la population pre-DGG. Les résultats obtenus en étendant des méthodes avancées d'inférence bayésienne, quantifiant les effets causaux directs, soutiennent l'hypothèse d'une influence directe de la susceptibilité aux mutations délétères dominantes sur la rétention des copies DGG. Ces résultats révèlent le mécanisme d'évolution non-adaptatif responsable de la rétention de gènes DGG susceptibles aux mutations délétères dominantes et notre extension de méthodes d'inférence bayesienne ouvre la voie à la quantification des relations causales directes dans un large ensemble de problématiquesGene families implicated in cancer and other genetic diseases have been greatly expanded through two rounds of whole-genome duplication (WGD) that occurred at the onset of jawed vertebrates. However, such gene duplicates are expected to lead to an enhanced susceptibility to genetic diseases, and thus their retention represents an evolutionary puzzle from a natural selection perspective. In this thesis, we have expanded classical population genetics models to reveal the non-adaptive mechanism through which such potentially deleterious ohnologs (WGD-duplicated genes) were retained in the vertebrate genomes. We have solved a deterministic haploid model, we have considered extensions to diploid genotypes, and we have analyzed population size effects and the impact of positive selection through a stochastic approach. The results demonstrate, consistently with available human cancer genome data, that ohnologs prone to dominant deleterious mutations are indirectly selected through purifying selection in post-WGD species, arisen through the ploidy incompatibility between post-WGD individuals and the rest of the pre-WGD population. Extending advanced Bayesian inference methods to quantify direct and indirect causal effects, we have found further supporting evidences for the direct role of the gene susceptibility to deleterious mutations on ohnolog retention. Our findings rationalize the evolutionary mechanism responsible for the expansion of ohnologs prone to dominant deleterious mutations, highlighting the role of WGD-induced speciation. Our extension of Bayesian inference methods paves the way for the identification of direct causal relationships in a huge variety of problems
28
Kabadi, Mohamed Ghassane. "Contribution à la Tolérance aux Défauts des Systèmes Complexes basée sur la Génération de Graphes Causaux." Thesis, Université de Lorraine, 2016. http://www.theses.fr/2016LORR0048/document.
Full textAbstract:
Le travail de thèse, qui s'inscrit dans le cadre du projet européen PAPYRUS (Plug and Play monitoring and control architecture for optimization of large scale production processes) du 7ème PCRD, a concerné tout d'abord la synthèse et la mise en œuvre d'une approche de modélisation, de diagnostic et de reconfiguration originale. Celle-ci se fonde sur la génération de graphes causaux permettant de modéliser en temps réel le comportement d'un système complexe dans un premier temps. La cible de cette première étude a été la papeterie Stora Enso d'Imatra en Finlande, qui était le procédé d'application du projet PAPYRUS. En suite logique à cette première partie, une approche permettant l'accommodation du système à certains défauts particuliers a été définie par l'ajustement des signaux de consigne de diverses boucles de régulation. Le manuscrit est structuré en trois parties. Dans la première, le projet européen PAPYRUS est présenté. Le rôle de chaque partenaire y est décrit au travers des différents « workpackages » et le travail de thèse y est positionné. La seconde partie de la thèse a pour objectif la génération d'un modèle utile au diagnostic en se fondant uniquement sur les différents signaux mesurés du système. Plus précisément, un modèle causal graphique est présenté par la mise en évidence des liens de causalité entre les différentes variables mesurées. Des analyses à base d'inter-corrélation, de transfert d'entropie et du test de causalité de Granger sont effectuées. Une approche de diagnostic fondée sur le modèle graphique ainsi obtenu est ensuite proposée en utilisant un test d'hypothèse séquentiel. La dernière partie est dédiée au problème d'accommodation aux défauts. Le graphe utilisé pour établir le diagnostic du système est remanié afin de faire apparaitre les différentes boucles de régulation du système. Une stratégie permettant la sélection de consignes influentes est alors proposée avec l'objectif d'ajuster ces dernières afin de compenser l'effet du défaut survenuThe thesis deals with modelling, diagnosis and fault tolerance of large scale processes. It is a part of the European project PAPYRUS (7th FWP). First, the European PAPYRUS project is described with details on the role of each partner through work package descriptions. The positioning of the thesis in this respect is also given. The second part of the thesis is about modelling graphical models for diagnosis purpose; more precisely, graphical causal model is used to highlight the causal links between the different variables of the process. To obtain such a model from data, several methods are proposed based on cross-correlation, entropy transfer and Granger causality. A diagnosis-based approach on the resulting graphical model is then proposed using statistical test and causal model of the process. This approach is illustrated using data from an industrial process and results are validated. The final section addresses fault tolerance based on digraph inferences and reference governor. This approach is illustrated using a MATLAB simulation which has a functional architecture similar to Board Machine 4 of Stora Enso IMATRA in Finland
29
Abdulaziz, Noor Amal Saud. "Evaluation of Texas Home Instruction for Parents of Preschool Youngsters Program on Reading and Math Achievement for Grades K to 8." Thesis, University of North Texas, 2019. https://digital.library.unt.edu/ark:/67531/metadc1538753/.
Full textAbstract:
This study was intended to evaluate the impact of socioeconomically disadvantaged children's participation in the Texas Home Instruction for Parents of Preschool Youngsters (TX HIPPY) Program on their school readiness and academic achievement. The study used a quasi-experimental design and applied full and optimal propensity score matching (PSM) to address the evaluation concern of the impact of the TX HIPPY program on HIPPY participants' academic achievement compared to non-HIPPY participants. This evaluation targeted former HIPPY participants and tracked them in the Dallas ISD database through Grade Levels K-8. Data were obtained by administering Istation's Indicators of Progress (ISIP) for kindergarten, TerraNova/SUPERA for Grades K-2, and State of Texas Assessments of Academic Readiness for math and reading (STAAR) for Grades 3-8. HIPPY and non-HIPPY groups were matched using propensity score analysis procedures. The evaluation findings show that the TX HIPPY program positively influences kindergarten students to start school ready to learn. The findings of math and reading achievements suggest that HIPPY children scored at the same level or higher than non-HIPPY children did on math and reading achievement, indicating that TX HIPPY program has achieved its goal of helping children maintain long-term academic success. However, the evaluation findings also indicated that the impact evaluation framework must be designed with attention to higher-level factors beyond academic achievement that influence children's academic success.
30
Pigache, François. "Modélisation causale en vue de la commande d'un translateur piézoélectrique plan pour une application haptique." Lille 1, 2005. https://pepite-depot.univ-lille.fr/LIBRE/Th_Num/2005/50376-2005-Pigache.pdf.
Full textAbstract:
Pour rendre compte physiquement de la manipulation d'un objet virtuel dans l'espace ou sur un plan, la plupart des dispositifs haptiques actuels font appel à des actionneurs à un seul degré de liberté, dont les actions sont couplées par diverses liaisons mécaniques (type pantographe). La technologie piézoélectrique est une solution avantageuse dans ce domaine d'utilisation, pour son important effort massique, le travail à faible vitesse, et surtout la capacité à motoriser plusieurs degrés de liberté à partir d'un seul actionneur. Pour cette raison, un translateur piézoélectrique plan à onde stationnaire est étudié. Un modèle simplifié est élaboré pour offrir une interprétation globale des phénomènes de contact. Il est établi selon le formalisme du graphe informationnel causal qui met en évidence deux asservissements applicables au domaine haptique :un retour d'effort actif par le contrôle en force, et une solution alternative comparable à un embrayage, qualifié de retour d'effort dissipatif.
31
Lessard, Pierre. "The application of computerized geographic information systems to epidemiological surveillance of cattle diseases caused by Theileria Parva." Diss., This resource online, 1990. http://scholar.lib.vt.edu/theses/available/etd-08032007-102231/.
Full text
32
Pigache, Francois. "Modélisation Causale en vue de la Commande d'un translateur piézoélectrique plan pour une application haptique." Phd thesis, Université des Sciences et Technologie de Lille - Lille I, 2005. http://tel.archives-ouvertes.fr/tel-00011938.
Full textAbstract:
Pour rendre compte physiquement de la manipulation d'un objet virtuel dans l'espace ou sur un plan, la plupart des dispositifs haptiques actuels font appel à des actionneurs à un seul degré de liberté, dont les actions sont couplées par diverses liaisons mécaniques (type pantographe). La technologie piézoélectrique est une solution avantageuse dans ce domaine d'utilisation, pour son important effort massique, le travail à faible vitesse, et surtout la capacité à motoriser plusieurs degrés de liberté à partir d'un seul actionneur. Pour cette raison, un translateur piézoélectrique plan à onde stationnaire est étudié. Un modèle simplifié est élaboré pour offrir une interprétation globale des phénomènes de contact. Il est établi selon le formalisme du graphe informationnel causal qui met en évidence deux asservissements applicables au domaine haptique : un retour d'effort actif par le contrôle en force, et une solution alternative comparable à un embrayage, qualifié de retour d'effort dissipatif.
33
Demersseman, Richard. "Caractérisation et modélisation causale d'un frein à liquide magnétorhéologique en vue de sa commande." Phd thesis, Paris, ENSAM, 2009. http://pastel.archives-ouvertes.fr/pastel-00005484.
Full textAbstract:
Les liquides magnétorhéologiques sont des suspensions de particules magnétiques micrométriques dans des liquides amagnétiques. Lorsqu'un tel liquide est exposé à un champ magnétique, les particules s'agrègent sous la forme de "chaînes" qui augmentent de façon importante la résistance à l'écoulement. Dans ce mémoire de thèse, on présente la conception d'un frein discoïde à liquide magnétorhéologique et sa caractérisation dans deux cas de fonctionnement. Dans le premier, l'axe décrit des triangles de vitesse lentement variables à courant constant, ce dernier étant varié à l'arrêt entre deux triangles. On a pu observer que le couple doit typiquement croître jusqu'à un seuil pour que la rotation s'amorce, puis qu'il "chute" avant de croître de nouveau tandis que la vitesse augmente. On a également remarqué que les seuils de couple mesurés aux premiers triangles après les variations du courant sont différents de ceux mesurés aux triangles suivants, qui se repètent. Une modélisation du frein a été proposée, valable uniquement pour le courant maximum admissible et dans le cas où l'axe a déjà décrit au moins un triangle de vitesse. Cette modélisation, élaborée en utilisant le formalisme Graphe Informationnel Causal (G.I.C.), permet de rendre compte de l'évolution du couple à l'amorçage de la rotation pour le premier triangle de vitesse, mais pas pour les suivants. Dans le second cas de fonctionnement, le frein est alimenté en courant lentement variable à vitesse constante. Différents relevés de l'hystérésis du couple en fonction du courant ont été obtenus. On a pu rendre compte précisément de ces derniers sur la base d'un modèle Eléments Finis 2D du frein et d'un modèle de comportement hystérétique pour l'acier.
34
Willardson, Spencer L. "Under the influence Of arms: the foreign policy causes and consequences of arms transfers." Diss., University of Iowa, 2013. https://ir.uiowa.edu/etd/2660.
Full textAbstract:
How are arms export choices made within a state? In this dissertation I use a foreign policy analysis framework to examine this question. I focus on examining each of the three primary levels of analysis in international relations as it relates to the main question. I begin with a typical international relations level and examine the characteristics of the two states that dominate the world arms trade: The United States and Russia. I then examine the full network of relations among all states in the international system that are involved in the sale or purchase of arms. To do this I use an Exponential Random Graph Model (ERGM) to examine these relations, which I derived from data on arms sales from the Stockholm International Peace Research Institute (SIPRI). I examine the arms sales in each decade from 1950 through 2010. In order to answer the question of how arms decisions are made within the state, I focus my inquiry on the United States and Russia. It is these states that have the practical capability to use arms transfers as a foreign policy tool. I examine the foreign policy making mechanisms in each of these states to determine how arms transfers can be used as a foreign policy tool. I examine and the bureaucratic institutions within each state and come up with a state ordering preference for how arms decisions are evaluated in each state. Finally, I use case studies to examine arms relations between the both the U.S. and Russia and three other states in each case. The other states were selected based on the pattern of sales between the two countries. I examine these sales to determine the impact of bureaucratic maneuvering and interest politics on the decision-making process within Russia and the United States. I find in my network analysis that the traditional measures of state power - military spending, regime type, and military alliances - do not account for the overall structure of the arms sale network. The most important factors in the formation of the arms sale network in each of the six decades that I study are specific configurations of triadic relations that suggest a continued hierarchy in the arms sale network. I find in my case study chapters that a simple model of state interest as a decision-making rule accounts for the decisions made by the different bureaucratic actors in the U.S. Russian arms sales are driven by a state imperative to increase Russia's market share, and there is high-level involvement in making different arms deals with other countries.
35
Williams, Mia Gabrielle. "Impact of environmental conditions on the infection behaviour of Western Australian strains of Plasmopara viticola, causal agent of downy mildew in grapevines." University of Western Australia. Soil Science and Plant Nutrition Discipline Group, 2006. http://theses.library.uwa.edu.au/adt-WU2006.0035.
Full textAbstract:
Downy mildew, caused by the biotrophic Oomycete Plasmopara viticola, is one of the most important diseases of grapevines world wide. It is particularly destructive in temperate viticultural regions that experience warm wet conditions during the vegetative growth of the vine (Wong et al., 2001). The disease is not normally a problem in mediterranean climates where the growing season tends to be hot and dry (Mullins et al., 1992; Sivasithamparam, 1993). Grape downy mildew is however a major disease in Australian viticulture (McLean et al., 1984; Magarey et al., 1991). Grape downy mildew was first reported in Europe in 1878 (Viennot-Bourgin, 1981). In Australia, it was recorded for the first time in 1917 at Rutherglen in Victoria (Vic) (de Castella, 1917). The first recorded outbreak of the disease in Western Australia (WA) occurred in 1997 in a small planting of vines in the far north of the state. In the subsequent year, it was detected in widespread commercial viticulture in the Swan Valley production area, near Perth (McKirdy et al., 1999). The pathogen has since been found in all grape growing regions of WA. Since its introduction into European vineyards in the 1880?s, P. viticola has become one of the world?s most investigated grapevine pathogens. Many aspects its basic biology however remain unknown (Wong et al., 2001). Due to the recent detection of P. viticola in WA, little is known of the nature of strains of the pathogen in the state and their response to local environmental conditions. Much of the research concerning the influence of environmental factors on the development of P. viticola has been conducted in Europe e.g. parts of France and Germany. Due to significant differences in climatic conditions and a shorter selection time on the pathogen in WA, much of the information described in European studies may not be directly applicable to the grape downy mildew disease situation in WA. The focus of this thesis was to examine epidemiological aspects of P. viticola in the mediterranean climate of WA. The environmental conditions that could favour the development of epidemics by strains of the pathogen that have been detected in the state were determined. The existence of P. viticola ecotypes and genetic variation among strains from WA and the Eastern states of Australia was also investigated.
36
Béarée, Richard. "Prise en compte des phénomènes vibratoires dans la génération de commande des machines-outils à dynamique élevée." Phd thesis, Paris, ENSAM, 2005. http://pastel.archives-ouvertes.fr/pastel-00001611.
Full textAbstract:
La recherche de performances toujours plus élevées conduit les systèmes de positionnement à dynamique élevée à la limite de leurs possibilités technologiques et structurelles, invalidant de ce fait l'hypothèse classiquement retenue d'une dynamique d'ensemble assimilable à celle d'un corps rigide. Il s'ensuit que la commande du système ne peut plus sous-estimer l'influence des phénomènes vibratoires sur la qualité du suivi de profil, mais doit moduler les efforts moteurs afin de réaliser l'adéquation entre les caractéristiques intrinsèques au système et les critères de rapidité et de précision escomptés. Les travaux développés dans cette étude visent une amélioration du comportement dynamique du système par une action sur la génération de commande. La génération de commande s'articule autour de deux principales fonctions: (1) l'élaboration d'une loi de mouvement pour la ou les variables du système à contrôler servant de référence pour les asservissements (commande indirecte) (2) la définition d'une précommande transformant la loi de mouvement en un signal de référence pour l'entrée du système physique (commande directe). Ces deux fonctions, correctement gérées, permettent de découpler les problèmes de régulation et de suivi de profil. Dans un premier temps, elles sont traitées spécifiquement pour le cas du mouvement d'un système monoaxe. Ainsi, l'influence de différentes classes de lois de mouvement (harmoniques, polynomiales, polynomiales par morceaux ou bang-bang) sur les vibrations et la durée du mouvement est formalisée. Une méthodologie d'élaboration d'une précommande adaptée aux souplesses d'un axe est présentée elle permet de mettre en exergue les contrainte.
37
Marano, Susan Aileen. "Smarticles: A Method for Identifying and Correcting Instability and Error Caused by Explicit Integration Techniques in Physically Based Simulations." DigitalCommons@CalPoly, 2014. https://digitalcommons.calpoly.edu/theses/1304.
Full textAbstract:
Using an explicit integration method in physically based animations has many advantages including conceptual and computational simplicity, however, it re- quires small time steps to ensure low numerical instability. Simulations with large numbers of individually interacting components such as cloth, hair, and fluid models, are limited by the sections of particles most susceptible to error. This results in the need for smaller time steps than required for the majority of the system. These sections can be diverse and dynamic, quickly changing in size and location based on forces in the system. Identifying and handling these trou- blesome sections could allow for a larger time step to be selected, while preventing a breakdown in the simulation.
This thesis presents Smarticles (smart particles), a method of individually de- tecting particles exhibiting signs of instability and stabilizing them with minimal adverse effects to visual accuracy. As a result, higher levels of error introduced from large time steps can be tolerated with minimal overhead. Two separate approaches to Smarticles were implemented. They attempt to find oscillating particles by analyzing a particle’s (1) past behavior and (2) behavior with re- spect to its neighbors along a strand. Both versions of Smarticles attempt to correct unstable particles using velocity dampening. Smarticles was applied to a two dimensional hair simulation modeled as a continuum using smooth particle hydrodynamic. Hair strands are formed by linking particles together using one of two methods: position based dynamics or mass-spring forces.
Both versions of Smarticles, as well as a control of normal particles, were directly compared and evaluated based on stability and visual fluidity. Hair particles were exposed to various forms of external forces under increasing time step lengths. Testing showed that both versions of Smarticles working together allowed an average increase of 18.62% in the time step length for hair linked with position based dynamics. In addition, Smarticles was able to significantly reduce visible instability at even larger time steps. While these results suggest Smarticles is successful, the method used to correct particle instability may jeopardize other important aspects of the simulation. A more accurate correction method would likely need to be developed to make Smarticles an advantageous method.
38
Mario, Justino Luiz. "Reação de híbridos de milho à podridão dos grãos causada por Stenocarpella macrospora e Stenocarpella maydis, em diferentes ambientes do Brasil." Universidade Federal de Uberlândia, 2010. https://repositorio.ufu.br/handle/123456789/15749.
Full textAbstract:
This work evaluated the reaction of 140 corn hybrids from double-haploid inbred derived from a crossing of a resistant inbred line with and susceptible inbred line. Being that, crossed with a conventional susceptible line which were inoculated with Stenocarpella maydis and Stenocarpella macrospora .The hybrids were assessed in three locations in the Miner Triangle and three locations in Southern Brazil to evaluate the difference of hybrids reaction to S. macrospora and S. maydis.
Two analysis were performed, one of the three locations in the Miner Triangle and one of the three locations in the South for the separation of five resistant hybrids, five average resistance and five susceptible to rot grain for the two locations. About the incidence of rot grain obtained from the hybrids, the quantification of S. macrospora, S.maydis and other fungi were done in the laboratory for the Miner Triangle and South of Brazil. In a joint analysis in the Miner Triangle it was found a frequency of 60% S. macrospora, 10% S.maydis and 30% other fungi. And in a joint analysis of South Brazil found 20% S. macrospora, 11% S.maydis and 69% of other fungi. The pathogen hybrid interaction showed the prevalence of S. macrospora in incidence of infected grains, with 60% in the Miner Triangle and 20% in the South, against S. maydis with 10% in the Miner Triangle and 11% in south of Brazil. These results showed that there were differences between the two places regarding the rotten grain caused by S. macrospora and S. maydis. Thus, positioning breeding programs in the pathosystem Stenocarpella-corn.Neste trabalho, avaliou-se a reação de 140 híbridos de milho provenientes de uma linhagem duplo-haplóide derivada do cruzamento, entre uma linhagem resistente e uma linhagem suscetível, cruzada com uma linhagem convencional suscetível. Os híbridos quais foram inoculados com Stenocarpella maydis (Berkeley) Sacc e Stenocarpella macrospora (Earle) sendo avaliados em três locais no Triângulo Mineiro e Três locais no Sul do Brasil. O objetivo foi identificar genótipos resistentes à S. macrospora e S. maydis e a sua quantificação nas duas regiões. Efetuou-se duas análises conjuntas, uma com os três locais do Triângulo Mineiro e outra com três locais do Sul verificar a reação dos híbridos quanto a resistência à grãos ardidos. Na incidência de grãos ardidos obtidos dos híbridos, realizou-se a quantificação de S. macrospora, S. maydis e outros fungos, para o Triângulo Mineiro e Sul. Na análise conjunta do Triângulo Mineiro ocorreu uma freqüência de 60% de S. macrospora, 10% de S. maydis e 30% de outros fungos. A análise conjunta dos dados do Sul obteve-se 20% de S.macrospora, 11% de S.maydis e 69% de outros fungos. A interação híbrido patógeno mostrou a prevalência da S. macrospora na incidência dos grãos infectados, com 60% no Triângulo Mineiro e de 20% no Sul e S. maydis com 10% no Triângulo Mineiro e 11% no Sul do Brasil. Esses resultados mostram que existem diferenças entre os dois locais, quanto à reação à podridão de grãos de milho causado por S. macrospora e S. maydis. Com isso pode-se direcionar os programas de melhoramento no patossistema Stenocarpella-milho.
Doutor em Genética e Bioquímica
39
Remy, Ghislain. "Commande optimisée d'un actionneur linéaire synchrone pour un axe de positionnement rapide." Phd thesis, Paris, ENSAM, 2007. http://tel.archives-ouvertes.fr/tel-00281529.
Full textAbstract:
Les moteurs linéaires sont devenus des composants incontournables dans le domaine de la conception des machines de production. L'actionneur synchrone à aimants permanents (PMLSM) constitue le principal composant et permet aujourd'hui d'obtenir des performances nettement supérieures à celles de son homologue rotatif accouplé à un système de transformation de mouvement. Cependant, pour utiliser ces constituants de façon optimale, il est nécessaire de prendre en compte dans la commande les spécificités de ce type d'actionneur. En effet, des phénomènes électriques et magnétiques, tels que les forces électromotrices non-sinusoïdales, la saturation des inductances et les forces de détente, génèrent des forces ondulantes perturbatrices pour la partie mécanique. L'objectif de cette étude est donc, dans un premier temps, d'améliorer la modélisation de ces moteurs linéaires en vue de leur commande. Des méthodes analytiques et éléments-finis sont utilisées pour quantifier l'importance de ces phénomènes sur la force de poussée. Puis, dans une deuxième partie, nous définissons les structures de commande optimales permettant de prendre en compte les phénomènes retenus. Le formalisme du Graphe Informationnel Causal (GIC), développé par le laboratoire L2EP, est utilisé pour représenter et définir de façon systématique les structures de commande retenues. Les méthodologies d'inversion du GIC sont détaillées pour permettre la définition des structures de commande en Boucles Fermées et Non-Fermées. Les structures classiques de commande industrielles sont analysées et de nouvelles structures de commande sont proposées pour la commande en force d'un moteur linéaire LMD10-050 de la société ETEL.
40
Mateus, Forero Andrea D. "DESIGN IN ADAPTATION TO DROUGHTS AND HEAT WAVES CAUSED BY CLIMATECHANGE IN RICE FARMS IN LERIDA, TOLIMA, COLOMBIA." Miami University / OhioLINK, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=miami151213167340772.
Full text
41
Lhomme, Serge. "Les réseaux techniques comme vecteur de propagation des risques en milieu urbain - Une contribution théorique et pratique à l'analyse de la résilience urbaine." Phd thesis, Université Paris-Diderot - Paris VII, 2012. http://tel.archives-ouvertes.fr/tel-00772204.
Full textAbstract:
Les réseaux techniques sont des infrastructures critiques qui peuvent être vulnérables aux inondations. Leur bon fonctionnement représente donc un enjeu majeur pour la gestion des risques et des territoires. Cette recherche postule que l'analyse de la résilience des réseaux techniques constitue un préalable nécessaire, mais non suffisant, à la mise en place de politiques d'amélioration de la résilience urbaine face aux inondations. Une méthodologie d'analyse de la résilience des réseaux techniques est proposée. Elle requiert l'analyse des capacités de résistance, d'absorption et de récupération de ces réseaux. Dans ce cadre, des développements ont été réalisés afin d'analyser les propriétés structurelles des réseaux techniques, les interdépendances entre les réseaux techniques et la remise en service de ces réseaux. Ces développements reposent principalement sur la théorie des graphes, les méthodes issues de la Sûreté de Fonctionnement, l'optimisation linéaire et l'analyse spatiale. Enfin, un prototype web-SIG a été développé puis appliqué à la ville de Dublin.
42
Blindu, Igor. "Outil d'aide au diagnostic du réseau d'eau potable pour la ville de Chisinau par analyse spatiale et temporelle des dysfonctionnements hydrauliques." Phd thesis, Ecole Nationale Supérieure des Mines de Saint-Etienne, 2004. http://tel.archives-ouvertes.fr/tel-00779032.
Full textAbstract:
Le travail effectué dans le cadre de cette thèse intitulée " Outil d'aide au diagnostic du réseau d'eau potable pour la ville de Chisinau par analyse spatiale et temporelle des dysfonctionnements hydrauliques " porte sur le développement d'une maquette du futur outil d'aide à la gestion des infrastructures et notamment du réseau d'eau potable de la ville de Chisinau Moldavie (1200 Km de canalisations - 800 000 habitants). La méthode proposée est basée sur l'analyse de l'état de fonctionnement du réseau d'eau potable. Cet état de fonctionnement du réseau d'AEP peut être connu à partir : - d'informations directes fournies par un système de télésurveillance (mesure de pression, de vitesse, de débit, de qualité....), - d'informations indirectes (analyse des incidents survenus sur le réseau, des interventions, de l'environnement du réseau....) obtenues. Dans notre cas, l'absence de mesures directes ne permet pas de quantifier l'état de fonctionnement du réseau sur l'ensemble du réseau sauf en quelques points critiques connus (station de pompage, station de relèvement..), c'est pourquoi, cet état est défini en se basant sur la liste des incidents, et des interventions survenues sur le réseau entre 1996 et 2001, ainsi que sur des informations portant sur l'environnement du réseau (nature des sols, aménagement du territoire ...) Ce travail de recherche comprend deux volets : Ü Aspect " Diagnostic " : Analyser qualitativement et quantitativement tous les aléas pouvant exister sur le réseau et se manifester par des observations. Il s'agit dans tous les cas d'établir le cheminement possible entre les observations, les causes possibles, et d'évaluer les conséquences induites. Il s'agit par une analyse successive et récursive (à l'aide de requêtes temporelles), de détecter la simultanéité de 2 ou plusieurs observations (manifestations de dysfonctionnement) se produisant dans un même laps de temps et la mise en évidence de relations topologiques et hydrauliques pouvant exister entre les sites où sont observés les dysfonctionnements. L'utilisation également de la théorie des graphes, plus particulièrement du réseau de Petri, permet de passer d'une analyse espace-temps entre 2 ou m événements à une analyse intégrant la causalité entre 2 événements. Ü Aspect " Aide à la décision " : Associer un " niveau d'urgence " à chaque tronçon du réseau afin d'assurer le suivi de la réhabilitation des infrastructures, l'assistance à la réhabilitation avec la détermination de zones prioritaires, la gestion/maintenance du réseau pour la pérennité du réseau. Ce niveau d'urgence est quantifié à l'aide d'une Méthode Hiérarchique Multicritères développée par SAATY (en considérant des critères techniques, économiques, sociaux, environnementaux ainsi que la politique des gestionnaires). La méthodologie développée utilise différents outils et méthodes issues : des bases de données temporelles, d'analyse spatiale et de SIG, de raisonnement cognitif et de modélisation hydraulique des écoulements, théorie de graphes et réseau de Petri. L'outil est testé sur un secteur pilote de la ville, qui représente environ 7% du réseau d'eau potable sur la ville, l'ensemble du réseau sera pris en compte ultérieurement lorsque la validation de cette portion de réseau sera faite par les services techniques de la ville de Chisinau (Moldavie). Mots clés : Vieillissement, réseau d'eau potable, Système d'Information Géographique, base de données géographique, renouvellement, méthode hiérarchique multicritère, dysfonctionnements, analyse spatio-temporelle, théorie des graphes, réseau de Petri, diagramme cause à effets.
43
Chen, Ming-Gue, and 陳銘菊. "Application of Directed Graphs to the Causal relationship between government Revenue and Expenditure." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/87631048566717773598.
Full textAbstract:
碩士逢甲大學
會計與財稅所
91
For these years, the expenditure of Taiwan government has been expanded drastically. Meanwhile, it was very difficult for the government to increase financial resource. As a result, the government’s fiscal deficit has been escalating. It demands our immediate attention to face the deficit problem. In this study, we intended to explore the empirical causal evidences between expenditure and revenue in Taiwan. We applied the directed graphs, which is very different from the traditional methods. Error correction model (ECM) analyzes the individual coefficient and the short-term movement construction. According to ECM’s innovation covariance chart, we can obtain the evolution of the causation at the same time, and can draw graphs to show the movement relation between the sequence to solve the problem of the covariance and the causation. The study showed that there are no relationships between Taiwan’s fiscal revenue and expenditure. There exists neither “tax-spend” nor “spend-tax” relationship. The government’s expanding expenditure and the government’s revenue are not related, which supports the independent hypothesis. The causations among expenditures are summarized as follows. “Expenditures for Community Development & Environmental Protection” is the effect of both “Expenditures Economic Development” and “Expenditures for Pension & Survivors’ Benefits”. On the other hand, “Expenditures Economic Development” and “Expenditures for Pension & Survivors’ Benefits” are the causes for “Expenditures for Community Development & Environmental Protection”. And “Expenditures for General Administration” is the effect for both “Obligations” and “Expenditures Economic Development”. In other words, “Obligations” and “Expenditures Economic Development” are the causes for “Expenditures for General Administration”. Hence, there exist causation relationships among expenditures.
44
Bernigau, Holger. "Causal Models over Infinite Graphs and their Application to the Sensorimotor Loop: Causal Models over Infinite Graphs and their Application to theSensorimotor Loop: General Stochastic Aspects and GradientMethods for Optimal Control." Doctoral thesis, 2014. https://ul.qucosa.de/id/qucosa%3A13254.
Full textAbstract:
Motivation and background
The enormous amount of capabilities that every human learns throughout his life, is probably among the most remarkable and fascinating aspects of life. Learning has therefore drawn lots of interest from scientists working in very different fields like philosophy, biology, sociology, educational sciences, computer sciences and mathematics. This thesis focuses on the information theoretical and mathematical aspects of learning.
We are interested in the learning process of an agent (which can be for example a human, an animal, a robot, an economical institution or a state) that interacts with its environment. Common models for this interaction are Markov decision processes (MDPs) and partially observable Markov decision processes (POMDPs). Learning is then considered to be the maximization of the expectation of a predefined reward function. In order to formulate general principles (like a formal definition of curiosity-driven learning or avoidance of unpleasant situation) in a rigorous way, it might be desirable to have a theoretical framework for the optimization of more complex functionals of the underlying process law. This might include the entropy of certain sensor values or their mutual information. An optimization of the latter quantity (also known as predictive information) has been investigated intensively both theoretically and experimentally using computer simulations by N. Ay, R. Der, K Zahedi and G. Martius. In this thesis, we develop a mathematical theory for learning in the sensorimotor loop beyond expected reward maximization.
Approaches and results
This thesis covers four different topics related to the theory of learning in the sensorimotor loop.
First of all, we need to specify the model of an agent interacting with the environment, either with learning or without learning. This interaction naturally results in complex causal dependencies. Since we are interested in asymptotic properties of learning algorithms, it is necessary to consider infinite time horizons. It turns out that the well-understood theory of causal networks known from the machine learning literature is not powerful enough for our purpose. Therefore we extend important theorems on causal networks to infinite graphs and general state spaces using analytical methods from measure theoretic probability theory and the theory of discrete time stochastic processes. Furthermore, we prove a generalization of the strong Markov property from Markov processes to infinite causal networks.
Secondly, we develop a new idea for a projected stochastic constraint optimization algorithm. Generally a discrete gradient ascent algorithm can be used to generate an iterative sequence that converges to the stationary points of a given optimization problem. Whenever the optimization takes place over a compact subset of a vector space, it is possible that the iterative sequence leaves the constraint set. One possibility to cope with this problem is to project all points to the constraint set using Euclidean best-approximation. The latter is
sometimes difficult to calculate. A concrete example is an optimization over the unit ball in a matrix space equipped with operator norm. Our idea consists of a back-projection using quasi-projectors different from the Euclidean best-approximation. In the matrix example, there is another canonical way to force the iterative sequence to stay in the constraint set:
Whenever a point leaves the unit ball, it is divided by its norm. For a given target function, this procedure might introduce spurious stationary points on the boundary. We show that this problem can be circumvented by using a gradient that is tailored to the quasi-projector used for back-projection. We state a general technical compatibility condition between a quasi-projector and a metric used for gradient ascent, prove convergence of stochastic iterative sequences and provide an appropriate metric for the unit-ball example.
Thirdly, a class of learning problems in the sensorimotor loop is defined and motivated. This class of problems is more general than the usual expected reward maximization and is illustrated by numerous examples (like expected reward maximization, maximization of the predictive information, maximization of the entropy and minimization of the variance of a given reward function). We also provide stationarity conditions together with appropriate gradient formulas.
Last but not least, we prove convergence of a stochastic optimization algorithm (as considered in the second topic) applied to a general learning problem (as considered in the third topic). It is shown that the learning algorithm converges to the set of stationary points. Among others, the proof covers the convergence of an improved version of an algorithm for the maximization of the predictive information as proposed by N. Ay, R. Der and K. Zahedi. We also investigate an application to a linear Gaussian dynamic, where the policies are encoded by the unit-ball in a space of matrices equipped with operator norm.
45
Pinho, Xavier Sá Castro. "Adverse outcome pathway for benzene induced toxicity through reverse causal reasoning and network perturbation analysis." Master's thesis, 2020. http://hdl.handle.net/10316/94047.
Full textAbstract:
Trabalho de Projeto do Mestrado Integrado em Engenharia Biomédica apresentado à Faculdade de Ciências e TecnologiaThe increase and improvement in molecular profiling technologies have enabled the acquisition of large datasets consisting of measurements for many molecular entities. These datasets allow an understanding of molecular profiles of, for example, a disease, drug and compounds action, or toxicity. Furthermore, gene expression profiling experiments usually produce extensive lists of differential expressed genes that characterize the comparison between the two states in the study, such as disease versus healthy or treatment versus control. In this study two approaches are used to interpret these lists, take out relevant and reliable hypotheses and quantify biological network perturbations: Reverse Causal Reasoning (RCR) and Network Perturbation Analysis (NPA); towards exploring the full potential of these datasets. The RCR and NPA methods are implemented and tested on the transcriptome of benzene-exposed individuals to propose a hypothesis of biological processes alterations. Several proposed altered biological mechanisms are in agreement with literature evidence, meaning that this approach can be a valuable tool for understanding mechanisms associated with benzene exposure. While some of them have not been studied and false positives are a possibility, this approach indicates possible candidates, that have not been verified by the literature as potential future directions in research.
The increase and improvement in molecular profiling technologies have enabled the acquisition of large datasets consisting of measurements for many molecular entities. These datasets allow an understanding of molecular profiles of, for example, a disease, drug and compounds action, or toxicity. Furthermore, gene expression profiling experiments usually produce extensive lists of differential expressed genes that characterize the comparison between the two states in the study, such as disease versus healthy or treatment versus control.In this study two approaches are used to interpret these lists, take out relevant and reliable hypotheses and quantify biological network perturbations: Reverse Causal Reasoning (RCR) and Network Perturbation Analysis (NPA); towards exploring the full potential of this datasets. The RCR and NPA methods are implemented and tested on the transcriptome of benzene-exposed individuals to propose a hypothesis of biological processes alterations.Several proposed altered biological mechanisms are in agreement with literature evidence, meaning that this approach can be a valuable tool for understanding mechanisms associated with benzene exposure. While some of them have not been studied and false positives are a possibility, this approach indicates possible candidates, that have not been verified by the literature as potential future directions in research.
46
-Yun, Sue, and 李淑芸. "On the Causal Realationship Among Taiwan’s Health Status、Health Expenditure、Labor Productivity and Income－An Application of Directed Graphs." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/05206575892036016468.
Full textAbstract:
碩士逢甲大學
財稅所
94
Every country’s health standard level cannot be separated from their annual income, which its presented one country’s economic development level. Every government should more make effort to build a better economic. We also acknowledge life quality,environment, medical resource and public health level is higher and the same time people more care about health. In fact, from one country macroeconomic analysis shows each income growth and development bring up the better supply of medical resource that affect people in the country will be healthier and bring up the labor productivity, push the people production up which all the make the economic developing. Because of the fact we can be sure that health, labor productivity and income is causal related to each other. To prove it, we are using Taiwan as the subject of our analysis. We collect the data from 1973 till 2003 and we are using time series and directed graph to analysis to find out cause and effect and the chain connection between health status, health expenditure, labor productivity and income. But, when we check cause and effect of the variable, the inspect traditional analysis only direct to few variable but can not check cause and effect from lot of variable in the same time so they use the directed graph to analysis this. The empirical result show, in the causal relation aspect, income can influence health significant and positive which means each time the income is grow higher, the health status is better, but the health expenditure is the negative and significant. These means people health status and OECD country’s health status is almost the same result. Life expectancy is increasing and infant mortality is decreasing, although the health expenditure is increasing, the health status is not positive influence. In directed graph analysis, health expenditure and income influence health status directly. Labor productivity has no influence in health status.
47
Hiran, Rahul. "Modeling Class of Software Vulnerabilities with Vulnerability Cause Graphs." Thesis, 2009. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-51854.
Full textAbstract:
Vulnerabilities discovered in software are not only due to programming errors but also due to designflaws. There are a number of methods to avoid design flaws which are all manual processes and needexpertise. We believe that the study of models of classes of vulnerabilities would give developerssufficient knowledge in how to avoid these vulnerabilities. A model of class of vulnerability can alsohelp in the decision making process during the software development process.In this thesis, we present a procedure for modeling a class of vulnerabilities given instances ofVulnerability Cause Graphs (VCGs). Using VCGs will structure the representation of causes tovulnerabilities.The approach presented in this thesis makes it possible to divide the work of modeling a class ofvulnerability without any permanent dependence on any specific persons. The approach is also flexible enough to accommodate new causes of vulnerabilities in software when being discovered.
48
Lai, Pei-Chun. "Causal Connection Search and Structural Demand Modeling on Retail-Level Scanner Data." Thesis, 2010. http://hdl.handle.net/1969.1/ETD-TAMU-2010-12-8922.
Full textAbstract:
Many researchers would be interested in one question: If a change of X is made, will Y be influenced in response? However, while a lot of statistical methods are developed to analyze association between variables, how to find a causal relationship among variables is relatively neglected.
The PC algorithm, developed on the basis of Pearl, Sprites, Glymour, and Scheines‟s studies, is used to find the causal pattern of the real-world observed data. However, PC in Tetrad produces a class of directed acyclic graphs (DAGs) that are statistically equivalent under a normal distribution, and therefore such a distributional assumption causes a series of unidentifiable DAGs because of the same joint probability.
In 2006 Shimizu, Hoyer, Hyvärinen, and Kerminen developed the Linear Independent Non-Gaussian Model (LiNGAM) to do a causal search based on the independently non-Gaussian distributed disturbances by applying higher-order moment structures. The research objective of this dissertation is to examine whether the LiNGAM is helpful relative to the PC algorithm, to detect the causal relation of non-normal data. The LiNGAM algorithm is implemented by first doing independent component analysis (ICA) estimation and then discovering the correct ordering of variables. Thus, the procedures of ICA estimation and the process of finding the correct causal orderings in LiNGAM are illustrated. Next, we do a causal search on the retail-level scanner data to investigate the pricing interaction between the manufacturer and the retailer by applying these two algorithms. While PC generates the set of indistinguishable DAGs, LiNGAM gives more exact causal patterns. This work demonstrates the algorithm based on the non-normal distribution assumption makes causal associations clearer. In Chapter IV, we apply a classical structural demand model to investigate the consumer purchase behavior in the carbonated soft drink market. Unfortunately, when further restrictions are imposed, we cannot get reasonable results as most researchers require. LiNGAM is applied to prove the existence of endogeneity for the brand‟s retail price and verify that the brand‟s wholesale price is not a proper instrument for its retail price. Therefore, consistent estimates cannot be derived as the theories suggest. These results imply that economic theory is not always found in restriction applied to observational data.
49
Bhattacharya, Indranil. "Feature Selection under Multicollinearity & Causal Inference on Time Series." Thesis, 2017. http://etd.iisc.ernet.in/2005/3980.
Full textAbstract:
In this work, we study and extend algorithms for Sparse Regression and Causal Inference problems. Both the problems are fundamental in the area of Data Science.
The goal of regression problem is to nd out the \best" relationship between an output variable and input variables, given samples of the input and output values. We consider sparse regression under a high-dimensional linear model with strongly correlated variables, situations which cannot be handled well using many existing model selection algorithms. We study the performance of the popular feature selection algorithms such as LASSO, Elastic Net, BoLasso, Clustered Lasso as well as Projected Gradient Descent algorithms under this setting in terms of their running time, stability and consistency in recovering the true support. We also propose a new feature selection algorithm, BoPGD, which cluster the features rst based on their sample correlation and do subsequent sparse estimation using a bootstrapped variant of the projected gradient descent method with projection on the non-convex L0 ball. We attempt to characterize the efficiency and consistency of our algorithm by performing a host of experiments on both synthetic and real world datasets.
Discovering causal relationships, beyond mere correlation, is widely recognized as a fundamental problem. The Causal Inference problems use observations to infer the underlying causal structure of the data generating process. The input to these problems is either a multivariate time series or i.i.d sequences and the output is a Feature Causal Graph where the nodes correspond to the variables and edges capture the direction of causality. For high dimensional datasets, determining the causal relationships becomes a challenging task because of the curse of dimensionality. Graphical modeling of temporal data based on the concept of \Granger Causality" has gained much attention in this context. The blend of Granger methods along with model selection techniques, such as LASSO, enables efficient discovery of a \sparse" sub-set of causal variables in high dimensional settings. However, these temporal causal methods use an input parameter, L, the maximum time lag. This parameter is the maximum gap in time between the occurrence of the output phenomenon and the causal input stimulus. How-ever, in many situations of interest, the maximum time lag is not known, and indeed, finding the range of causal e ects is an important problem. In this work, we propose and evaluate a data-driven and computationally efficient method for Granger causality inference in the Vector Auto Regressive (VAR) model without foreknowledge of the maximum time lag. We present two algorithms Lasso Granger++ and Group Lasso Granger++ which not only constructs the
hypothesis feature causal graph, but also simultaneously estimates a value of maxlag (L) for each variable by balancing the trade-o between \goodness of t" and \model complexity".
50
Zhao, Ke. "Development of a data driven multiple observer and causal graph approach for fault diagnosis of nuclear power plant sensors and field devices /." 2002. http://etd.utk.edu/2002/ZhaoKe.pdf.
Full textAbstract:
Thesis (M.S.)--University of Tennessee, Knoxville, 2002.Title from title page screen (viewed on Feb. 28, 2003). Thesis advisor: B. R. Upadhyaya. Document formatted into pages (xiii, 179 p. : ill. (chiefly col.)). Vita. Includes bibliographical references (p.142-147).