- / -

Τα συστήματα διεπαφών ανθρώπινου εγκεφάλου-υπολογιστή (BCIs: Brain-Computer Interfaces) απαιτούν την πραγματικού χρόνου, αποτελεσματική επεξεργασία των μετρήσεων των ηλεκτροεγκεφαλογραφικών (ΗΕΓ) σημάτων του χρήστη τους, προκειμένου να μεταφράσουν τις νοητικές διεργασίες/προθέσεις του σε σήματα ελέγχου εξωτερικών διατάξεων ή συστημάτων. Στο πλαίσιο της εργασίας αυτής μελετήθηκε το θεωρητικό υπόβαθρο του προβλήματος και αναλύθηκαν συνοπτικά οι κυριότερες τεχνικές που χρησιμοποιούνται σήμερα. Επιπρόσθετα, παρουσιάστηκε μία μέθοδος ταξινόμησης των νοητικών προθέσεων της αριστερής και δεξιάς κίνησης των χεριών ενός χρήστη η οποία εφαρμόστηκε σε πραγματικά ιατρικά δεδομένα. Η εξαγωγή των χαρακτηριστικών που διαφοροποιούνται μεταξύ των δύο καταστάσεων βασίστηκε σε πληροφορίες του πεδίου χρόνου-συχνότητας, οι οποίες αντλούνται με το φιλτράρισμα των ακατέργαστων ΗΕΓ δεδομένων και με τη βοήθεια των αιτιατών κυματιδίων Morlet, ενώ για την επακόλουθη ταξινόμηση των χαρακτηριστικών αναπτύχθηκαν και συγκρίθηκαν δύο αξιόπιστες μέθοδοι. Η πρώτη αφορά στη δημιουργία πιθανοθεωρητικών προτύπων κανονικής κατανομής για κάθε κατηγορία πρόθεσης κίνησης, με την τελική απόφαση ταξινόμησης να λαμβάνεται με εφαρμογή του απλού ταξινομητή του Bayes, ενώ η δεύτερη δημιουργεί ένα πρότυπο ταξινόμησης με βάση το θεωρητικό πλαίσιο των Μηχανών Διανυσμάτων Υποστήριξης (SVM). Στόχος του προβλήματος της δυαδικής ταξινόμησης είναι να αποφασίζεται σε ποια από τις δύο κατηγορίες ανήκει μία δεδομένη νοητική πρόθεση όσο το δυνατόν ταχύτερα και αξιόπιστα, έτσι ώστε ο σχεδιαζόμενος αλγόριθμος να εξυπηρετήσει ένα πλαίσιο ανατροφοδότησης της τελικής απόφασης στο χρήστη σε συνθήκες πραγματικού χρόνου. / Brain-Computer Interfaces (BCIs) demand the efficient processing of EEG data in order to translate one's thought or wish into a control signal that can be applied as input to external devices. Here we present a method to classify left from right hand movements, by extracting features from the data with Morlet wavelets and classifying with two different models, SVMs and Naive Bayes Classifier.

As autonomous driving is rapidly becoming the next major challenge in the auto- motive industry, the problem of Simultaneous Localization And Mapping (SLAM) has never been more relevant than it is today. This thesis presents the idea of examining SLAM algorithms by implementing such an algorithm on a radio con- trolled car which has been fitted with sensors and microcontrollers. The software architecture of this small-scale vehicle is based on the Robot Operating System (ROS), an open-source framework designed to be used in robotic applications. This thesis covers Extended Kalman Filter (EKF)-based SLAM, FastSLAM, and GraphSLAM, examining these algorithms in both theoretical investigations, simulations, and real-world experiments. The method used in this thesis is model- based development, meaning that a model of the vehicle is first implemented in order to be able to perform simulations using each algorithm. A decision of which algorithm to be implemented on the physical vehicle is then made backed up by these simulation results, as well as a theoretical investigation of each algorithm. This thesis has resulted in a dynamic model of a small-scale vehicle which can be used for simulation of any ROS-compliant SLAM-algorithm, and this model has been simulated extensively in order to provide empirical evidence to define which SLAM algorithm is most suitable for this application. Out of the algo- rithms examined, FastSLAM was proven to the best candidate, and was in the final stage, through usage of the ROS package gMapping, successfully imple- mented on the small-scale vehicle.

In a wide range of applications, decisions must be made by combining information from multiple agents with varying levels of trust and expertise. For example, citizen science involves large numbers of human volunteers with differing skills, while disaster management requires aggregating information from multiple people and devices to make timely decisions. This thesis introduces efficient and scalable Bayesian inference for decision combination, allowing us to fuse the responses of multiple agents in large, real-world problems and account for the agents’ unreliability in a principled manner. As the behaviour of individual agents can change significantly, for example if agents move in a physical space or learn to perform an analysis task, this work proposes a novel combination method that accounts for these time variations in a fully Bayesian manner using a dynamic generalised linear model. This approach can also be used to augment agents’ responses with continuous feature data, thus permitting decision-making when agents’ responses are in limited supply. Working with information inferred using the proposed Bayesian techniques, an information-theoretic approach is developed for choosing optimal pairs of tasks and agents. This approach is demonstrated by an algorithm that maintains a trustworthy pool of workers and enables efficient learning by selecting informative tasks. The novel methods developed here are compared theoretically and empirically to a range of existing decision combination methods, using both simulated and real data. The results show that the methodology proposed in this thesis improves accuracy and computational efficiency over alternative approaches, and allows for insights to be determined into the behavioural groupings of agents.

Le rendu photo-réaliste traditionnel exige un effort manuel et des calculs intensifs pour créer des scènes et rendre des images réalistes. C'est principalement pour cette raison que la création de contenus pour l’imagerie numérique de haute qualité a été limitée aux experts et le rendu hautement réaliste nécessite encore des temps de calcul significatifs. Le rendu basé image (IBR) est une alternative qui a le potentiel de rendre les applications de création et de rendu de contenus de haute qualité accessibles aux utilisateurs occasionnels, puisqu'ils peuvent générer des images photo-réalistes de haute qualité sans subir les limitations mentionnées ci-dessus. Nous avons identifié trois limitations importantes des méthodes actuelles de rendu basé image : premièrement, chaque algorithme possède des forces et faiblesses différentes, en fonction de la qualité de la reconstruction 3D et du contenu de la scène, et un seul algorithme ne permet souvent pas d’obtenir la meilleure qualité de rendu partout dans l’image. Deuxièmement, ces algorithmes présentent de forts artefacts lors du rendu d’objets manquants ou partiellement reconstruits. Troisièmement, la plupart des méthodes souffrent encore d'artefacts visuels significatifs dans les régions de l’image où la reconstruction est de faible qualité. Dans l'ensemble, cette thèse propose plusieurs améliorations significatives du rendu basé image aussi bien en termes de vitesse de rendu que de qualité d’image. Ces nouvelles solutions sont basées sur le rendu sélectif, la substitution de modèle basé sur l'apprentissage, et la prédiction et la correction des erreurs de profondeur. / Traditional photo-realistic rendering requires intensive manual and computational effort to create scenes and render realistic images. Thus, creation of content for high quality digital imagery has been limited to experts and highly realistic rendering still requires significant computational time. Image-Based Rendering (IBR) is an alternative which has the potential of making high-quality content creation and rendering applications accessible to casual users, since they can generate high quality photo-realistic imagery without the limitations mentioned above. We identified three important shortcomings of current IBR methods: First, each algorithm has different strengths and weaknesses, depending on 3D reconstruction quality and scene content and often no single algorithm offers the best image quality everywhere in the image. Second, such algorithms present strong artifacts when rendering partially reconstructed objects or missing objects. Third, most methods still result in significant visual artifacts in image regions where reconstruction is poor. Overall, this thesis addresses significant shortcomings of IBR for both speed and image quality, offering novel and effective solutions based on selective rendering, learning-based model substitution and depth error prediction and correction.

The aim of this study is to evaluate feasibility of using Bayes classifiers for predicting corporate bankruptcies. The results obtain show that Bayes classifiers do reach comparable results to then more commonly used methods such the logistic regression and the decision trees. The comparison has been carried out based on Czech and Polish data sets. The overall accuracy rate of these so called naive Bayes classifiers, using entropic discretization along with the hybrid pre-selection of the explanatory attributes, reaches 77.19 % for the Czech dataset and 79.76 % for the Polish set respectively. The AUC values for these data sets are 0.81 and 0.87. The results obtained for the Polish data set have been compared to the already published articles by Tsai (2009) and Wang et al. (2014) who applied different classification algorithms. The method proposed in my study, when compared to the above earlier works, comes out as quite successful. The thesis also includes comparing various approaches as regards the discretisation of numerical attributes and selecting the relevant explanatory attributes. These are the key issues for increasing performance of the naive Bayes classifiers

O objetivo principal do trabalho foi o desenvolvimento de um programa computacional, em linguagem Microsoft Visual Basic 6.0 (versão executável), para estimativa da taxa de penetrância a partir da análise de genealogias com casos de doenças com herança autossômica dominante. Embora muitos dos algoritmos empregados no programa tenham se baseado em idéias já publicadas na literatura (em sua maioria por pesquisadores e pós-graduandos do Laboratório de Genética Humana do Instituto de Biociências da Universidade de São Paulo), desenvolvemos alguns métodos inéditos para lidar com situações encontradas com certa frequência nos heredogramas publicados na literatura, como: a) ausência de informações sobre o fenótipo do indivíduo gerador da genealogia; b) agrupamento de árvores de indivíduos normais sem a descrição da distribuição de filhos entre os progenitores; c) análise de estruturas da genealogia contendo uniões consanguíneas, utilizando um método alternativo ao descrito na literatura; d) determinação de soluções gerais para as funções de verossimilhança de árvores de indivíduos normais com ramificação regular e para as probabilidades de heterozigose de qualquer indivíduo pertencente a essas árvores. Além da versão executável, o programa, denominado PenCalc, é apresentado também numa versão para Internet (PenCalc Web), a qual fornece adicionalmente as probabilidades de heterozigose e o cálculo de afecção na prole de todos os indivíduos da genealogia. Essa versão pode ser acessada livre e gratuitamente no endereço http://www.ib.usp.br/~otto/pencalcweb. Desenvolvemos também um modelo com taxa de penetrância variável dependente da geração, uma vez que a inspeção de famílias com doenças autossômicas dominantes, como é o caso da síndrome da ectrodactilia associada à hemimelia tibial (EHT), sugere a existência de um fenômeno similar à antecipação, em relação à taxa de penetrância. Os modelos com taxa de penetrância constante e variável, e os métodos desenvolvidos neste trabalho foram aplicados a 21 heredogramas de famílias com afetados pela EHT e ao conjunto das informações de todas essas genealogias (meta-análise), obtendo-se em todos os casos estimativas da taxa de penetrância. / The main objective of this dissertation was the development of a computer program, in Microsoft® Visual Basic® 6.0, for estimating the penetrance rate of autosomal dominant diseases by means of the information contained on genealogies. Some of the algorithms we used in the program were based on ideas already published in the literature by researchers and (post-) graduate students of the Laboratory of Human Genetics, Department of Genetics and Evolutionary Biology, Institute of Biosciences, University of São Paulo. We developed several other methods to deal with particular structures found frequently in the genealogies published in the literature, such as: a) the absence of information on the phenotype of the individual generating of the genealogy; b) the grouping of trees of normal individuals without the separate description of the offspring number per individual; c) the analysis of structures containing consanguineous unions; d) the determination of general solutions in simple analytic form for the likelihood functions of trees of normal individuals with regular branching and for the heterozygosis probabilities of any individual belonging to these trees. In addition to the executable version of the program summarized above, we also prepared, in collaboration with the dissertation supervisor and the undergraduate student Marcio T. Onodera (main author of this particular version), another program, represented by a web version (PenCalc Web). It enables the calculation of heterozygosis probabilities and the offspring risk for all individuals of the genealogy, two details we did not include in the present version of our program. The program PenCalc Web can be accessed freely at the home-page address http://www.ib.usp.br/~otto/pencalcweb. Another important contribution of this dissertation was the development of a model of estimation with generationdependent penetrance rate, as suggested by the inspection of families with some autosomal dominant diseases, such as the ectrodactyly-tibial hemimelia syndrome (ETH), a condition which exhibits a phenomenon similar to anticipation in relation to the penetrance rate. The models with constant and variable penetrance rates, as well as practically all the methods developed in this dissertation, were applied to 21 individual genealogies from the literature with cases of ETH and to the set of all these genealogies (meta-analysis). The corresponding results of all these analysis are comprehensively presented.

Orientador: Luís Cuadrado Martin / Resumo: O conhecimento sobre qualquer patologia pode ser facilmente encontrado na internet, mas dificilmente encontra-se alguma ferramenta que faça a análise e o raciocínio entre os dados de um paciente e se obtenha o diagnóstico mais provável. Em nosso cotidiano, em virtude de uma maior demanda na área da saúde, existe uma necessidade crescente de diagnósticos médicos rápidos e precisos. Em virtude disso, foi elaborado um Sistema de Apoio à Decisão Médica com o intuito de otimizar e agilizar de forma confiável os diagnósticos médicos. A ideia é dar qualidade e agilidade à prática médica, adotando a tecnologia como ferramenta básica: “Quem tem mais informação, tem melhores condições para escolher e tomar decisões”. Na construção deste sistema, foram utilizados um banco de dados relacional (MySQL) e aplicadas técnicas de inteligência artificial, tais como: a construção de Árvores de Decisão, Aprendizado não supervisionado e a utilização das Redes de Bayes (onde estão envolvidos domínios de conhecimento com significativo grau de incerteza, como é o caso da área médica). Através da união destas técnicas, são feitas a seleção e classificação das doenças mais prováveis, onde as mesmas podem ser examinadas com mais detalhes pelo médico, garantindo assim uma maior segurança na escolha dos possíveis diagnósticos. Visando uma maior abrangência e rapidez na disseminação do conhecimento humano, o sistema foi disponibilizado via internet (www.danton.med.br). Para a concepção do projeto foi reali... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: The knowledge about any pathology can be easily found on the internet, but it is difficult to find any tool that makes the analysis and reasoning between the data of a patient and obtain the most probable diagnosis. In our daily lives, due to a greater demand in the health area, there is a growing need for fast and accurate medical diagnoses. As a result, a Medical Decision Support System was developed in order to reliably optimize and streamline medical diagnostics. The idea is to give quality and agility to medical practice, adopting technology as a basic tool: “Who has more information, has better conditions to choose and make decisions”. In the construction of this system, a relational database (MySQL) was used and artificial intelligence techniques were applied, such as: the construction of Decision Trees, Unsupervised Learning and the use of Bayes Networks (where knowledge domains are involved with significant degree of uncertainty, as is the case in the medical field). Through the union of these techniques, the selection and classification of the most probable diseases are made, where they can be examined in more detail by the doctor, thus ensuring greater security in the choice of possible diagnoses. Aiming at a greater scope and speed in the dissemination of human knowledge, the system was made available via internet (www.danton.med.br). To design the project, a prospective, randomized, crossover and open study was carried out; in which 3 groups of doctors (called gr... (Complete abstract click electronic access below) / Doutor

Für die diagnostische Inferenz unter Unsicherheit werden Bayessche Netze untersucht. Grundlage dafür bildet eine adäquate einheitliche Repräsentation des notwendigen Wissens. Dies ist sowohl generisches als auch auf Erfahrungen beruhendes spezifisches Wissen, welches in einer Wissensbasis gespeichert wird. Zur Wissensverarbeitung wird eine Kombination der Problemlösungsmethoden des Concept Based und Case Based Reasoning eingesetzt. Concept Based Reasoning wird für die Diagnose-, Therapie- und Medikationsempfehlung und -evaluierung über generischesWissen eingesetzt. Sonderfälle in Form von spezifischen Patientenfällen werden durch das Case Based Reasoning verarbeitet. Darüber hinaus erlaubt der Einsatz von Bayesschen Netze den Umgang mit Unsicherheit, Unschärfe und Unvollständigkeit. Es können so die gültigen allgemeinen Konzepte nach derenWahrscheinlichkeit ausgegeben werden. Dazu werden verschiedene Inferenzmechanismen vorgestellt und anschließend im Rahmen der Entwicklung eines Prototypen evaluiert. Mit Hilfe von Tests wird die Klassifizierung von Diagnosen durch das Netz bewertet.:1 Einleitung 2 Repräsentation und Inferenz 3 Inferenzmechanismen 4 Prototypische Softwarearchitektur 5 Evaluation 6 Zusammenfassung

Le comportement moteur est un phénomène où les différentes composantes d’un système biologique sont organisées de façon à assurer la coordination d’un mouvement intentionnel. Selon les théories computationnelles, le comportement est défini comme un problème moteur dont la solution peut être trouvée par des systèmes divisés de manière hiérarchique. Les composantes traitent et communiquent entre elles de l’information représentant les aspects pertinents du problème moteur (positions, trajectoires, vitesses, forces, etc.) lesquels sont censés être organisés à leur tour selon une hiérarchie d’abstraction et de complexité ascendante. Le défi est de faire face à quatre problèmes centraux du comportement : a) Le nombre élevé de degrés de liberté et d’interactions ; b) La redondance des degrés de liberté ; c) L’anticipation des effets du mouvement ; d) L’incertitude dans l’information. Les théories computationnelles classiques proposent des schémas explicatifs composés d’un agencement de différents modèles internes (prospectifs et inverses). Plus récemment, l’approche bayésienne propose un schéma hiérarchique plus homogène lequel est censé faire face aussi à l’incertitude de l’information. Cette recherche démontre que les théories computationnelles, y compris l’approche bayésienne, sont paralysées par un dilemme insurmontable : soit elles peuvent passer à l’échelle de manière computationnellement faisable - les calculs peuvent être réalisés en un temps raisonnable - mais dans ce cas elles ne peuvent pas reproduire la versatilité caractéristique du comportement des êtres vivants ; soit elles aspirent à reproduire la versatilité biologique mais alors elles sont infaisables. / Motor behavior is a phenomenon where the components making up a biological system are organized so as to ensure the coordination of a purposeful movement. According to computational theories, behavior is defined as a motor problem the solution of which can be found by systems divided hierarchically. The components process and communicate information representing the relevant variables of the motor problem (positions, trajectories, velocities, forces, etc.) which are, in turn, assumed to be organized as a hierarchy of increasing abstraction and complexity. The challenge is to tackle the four core problems of behavior: a) The high number of degrees of freedom and their interactions; b) The redundancy of degrees of freedom; c) The anticipation of the effects of movement; d) The uncertainty in information. Classical computational theories advance explanatory schemas made of structured sets of internal models (forward and inverse). More recently, the Bayesian approach advances a more homogeneous hierarchical schema which is supposed to account for uncertainty in information. This research shows that computational theories, including the Bayesian approach, are crippled by an unsolvable dilemma: The first horn is that if the models can scale up while staying computationally tractable, i.e. the computations can be carried out in a reasonable amount of time, then they fail to reproduce the versatility which characterizes the behavior of living beings. The second horn is that if the models aspire to reproduce biological versatility, then they are intractable.