Pathway is a set of genes which are predefined and serve a particular celluar or physiological function. Ranking pathways relevant to a particular phenotype can help researchers focus on a few sets of genes in pathways. In this thesis, a Bayesian hierarchical latent model was proposed using generalized linear random effects model. The advantage of the approach was that it can easily incorporate prior knowledges when the sample size was small and the number of genes was large. For the covariance matrix of a set of random variables, two Gaussian random processes were considered to construct the dependencies among genes in a pathway. One was based on the polynomial kernel and the other was based on the Gaussian kernel. Then these two kernels were compared with constant covariance matrix of the random effect by using the ratio, which was based on the joint posterior distribution with respect to each model. For mixture models, log-likelihood values were computed at different values of the mixture proportion, compared among mixtures of selected kernels and point-mass density (or constant covariance matrix). The approach was applied to a data set (Mootha et al., 2003) containing the expression profiles of type II diabetes where the motivation was to identify pathways that can discriminate between normal patients and patients with type II diabetes. / Master of Science

La méthode que nous présentons pour modéliser des données dites de "comptage" ou données de Poisson est basée sur la procédure nommée Modélisation multi-niveau et interactive de la régression de Poisson (PRIMM) développée par Christiansen et Morris (1997). Dans la méthode PRIMM, la régression de Poisson ne comprend que des effets fixes tandis que notre modèle intègre en plus des effets aléatoires. De même que Christiansen et Morris (1997), le modèle étudié consiste à faire de l'inférence basée sur des approximations analytiques des distributions a posteriori des paramètres, évitant ainsi d'utiliser des méthodes computationnelles comme les méthodes de Monte Carlo par chaînes de Markov (MCMC). Les approximations sont basées sur la méthode de Laplace et la théorie asymptotique liée à l'approximation normale pour les lois a posteriori. L'estimation des paramètres de la régression de Poisson est faite par la maximisation de leur densité a posteriori via l'algorithme de Newton-Raphson. Cette étude détermine également les deux premiers moments a posteriori des paramètres de la loi de Poisson dont la distribution a posteriori de chacun d'eux est approximativement une loi gamma. Des applications sur deux exemples de données ont permis de vérifier que ce modèle peut être considéré dans une certaine mesure comme une généralisation de la méthode PRIMM. En effet, le modèle s'applique aussi bien aux données de Poisson non stratifiées qu'aux données stratifiées; et dans ce dernier cas, il comporte non seulement des effets fixes mais aussi des effets aléatoires liés aux strates. Enfin, le modèle est appliqué aux données relatives à plusieurs types d'effets indésirables observés chez les participants d'un essai clinique impliquant un vaccin quadrivalent contre la rougeole, les oreillons, la rub\'eole et la varicelle. La régression de Poisson comprend l'effet fixe correspondant à la variable traitement/contrôle, ainsi que des effets aléatoires liés aux systèmes biologiques du corps humain auxquels sont attribués les effets indésirables considérés. / We propose a method for analysing count or Poisson data based on the procedure called Poisson Regression Interactive Multilevel Modeling (PRIMM) introduced by Christiansen and Morris (1997). The Poisson regression in the PRIMM method has fixed effects only, whereas our model incorporates random effects. As well as Christiansen and Morris (1997), the model studied aims at doing inference based on adequate analytical approximations of posterior distributions of the parameters. This avoids the use of computationally expensive methods such as Markov chain Monte Carlo (MCMC) methods. The approximations are based on the Laplace's method and asymptotic theory. Estimates of Poisson mixed effects regression parameters are obtained through the maximization of their joint posterior density via the Newton-Raphson algorithm. This study also provides the first two posterior moments of the Poisson parameters involved. The posterior distributon of these parameters is approximated by a gamma distribution. Applications to two datasets show that our model can be somehow considered as a generalization of the PRIMM method since it also allows clustered count data. Finally, the model is applied to data involving many types of adverse events recorded by the participants of a drug clinical trial which involved a quadrivalent vaccine containing measles, mumps, rubella and varicella. The Poisson regression incorporates the fixed effect corresponding to the covariate treatment/control as well as a random effect associated with the biological system of the body affected by the adverse events.

Consideramos dois tópicos distintos relacionados a modelos clássicos da mecânica estatísticas de equilíbrio. O primeiro constitui-se na análise de equação parabólicas semi-lineares associadas à transformação de grupo de renormalização para o gás de Coulomb hierárquico bidimensional e o gás dipolos hierárquicos em dimensão d>1 após tomarmos um limite apropriado (limite L 1 do tamanho do bloco). O outro tópico estudado foi a construção de uma função majorante (, z) para a pressão termodinâmica de um gás formado por partículas interagentes com atividade z e temperatura -1, cuja interação entre dois corpos pode ser decomposta em escalas como um potencial estável. Somos capazes de demonstrar que o problema de valor inicial dado pela equação do gás de Coulomb está bem definido (existência, unicidade e dependência contínua das soluções) em um espaço funcional adequado e a solução converge assintoticamente para uma das infinitas contáveis soluções de equilíbrio. Quanto ao gás de dipolos, embora não tenhamos conseguido provar a existência e unicidade das soluções, garantimos que a única solução estacionária limitada inferiormente é a trivial nula, que é uma solução estável. Ao menos no caso dos modelos hierárquicos, os resultados obtidos permitem dar uma resposta definitiva à conjectura de Gallavotti e Nicolò sobre uma sequência infinita de transições de fase. A função majorante é construída como a solução de uma equação diferencial parcial quase-linear de primeira ordem. Através da do método das características relacionamos a solução (majorante) à função W de Lambert cuja expansão em série possui uma singularidade originada pelo corte que a função W possui no plano complexo. A descrição da função majorante como uma função W possui no plano complexo. A descrição da função majorante como uma função W permite uma melhora nas estimativas de raio de convergência para série de Mayer para pressão. / We have considered in this thesis two distinct topics related to classic models in equilibrium statistical mechanics. The first one is the analysis of semilinear parabolic partial differential equations given by a suitable limit (size of block L 1) in the renormalization group for the dipole gas in any dimension d>1. The other topic is the construction of a majorant function (, z) for the thermodynamic -1 whose potential admits a scale decomposition in terms of some stable potential. We are capable to demonstrate the well-posedness (existence, uniqueness and continuous dependence of solutions) for Coulomb gas equations and the global asymptotic convergence of the flow to one of its countably many equilibrium solutions. The dipole gas equations are technically more difficult and lack the results weve achieved in Coulomb gas but, despite its difficulties, we can establish the uniqueness of the trivial solution as a equilibrium ane and its stabilish. At least for hierarchical models, the established results give a definite answer to Gallovotti and Niclolòs conjecture of na infinite of phase transitions. The majorant function is constructed as the solution of a first order quase-linear partial differential equation. By means of the characteristics method we are able to relate its solution (the majorant) to Lamberts W-function whose series expansion possess a singularity given by W-function allows better estimates for Mayer series convergence.

A observação de gradientes latitudinais em aspectos da história de vida de aves tem motivado o estudo da evolução e variabilidade das histórias de vida nestes organismos. Um exemplo bem documentado é a variação no tamanho da ninhada, onde aves de latitudes menores tendem a ter ninhadas menores do que os seus homólogos de latitudes altas. Uma hipótese que visa explicar esta variação propõe que a sobrevivência em latitudes tropicais é maior para compensar o tamanho da ninhada menor e evitar a extinção das populações. Esta explicação tem tido grande aceitação e apoio por parte de alguns estudos, mas tem sido questionada por outros que não encontraram taxas de sobrevivência mais elevadas em aves tropicais. De modo implícito, todos estes estudos basearam seus resultados na sobrevivência de indivíduos adultos. As populações com o tamanho da ninhada menor não poderiam crescer da mesma maneira que as populações com ninhadas maiores; portanto, se justifica acreditar que algo deve mudar com a latitude para manter o balanço em tamanho populacional. Na busca por explicações alternativas para a persistência das populações de aves tropicais com relativamente pequenos tamanhos de ninhada, surge outra hipótese que propõe que, se não houver diferenças na sobrevivência de indivíduos adultos entre latitudes, o aspecto fundamental que varia é a sobrevivência juvenil, com sobrevivência maior para os juvenis das zonas tropicais em comparação com os juvenis das zonas temperadas. No entanto, atualmente há pouca evidência que suporta esta conclusão. Os resultados contrastantes desses estudos sugerem a falta de um consenso geral sobre a hipótese de que as aves tropicais têm taxas de sobrevivência mais elevadas do que as aves de regiões temperadas, motivando a formulação de hipóteses alternativas e convidando novos testes de hipótese. Neste estudo, pretendemos a) avaliar o efeito da idade sobre a sobrevivência em aves tropicais, estimando as probabilidades anuais de sobrevivência aparentes idade-específicas para um conjunto de aves passeriformes de sub-bosque na Amazônia central brasileira; e b) contribuir para o debate sobre o gradiente latitudinal na sobrevivência de adultos, comparando nossas estimativas com estimativas de outras latitudes. Para estimar a sobrevivência idade-específica ajustamos aos nossos dados um modelo Cormack-Jolly-Seber (CJS) hierárquico para n espécies, que trata os parâmetros espécie-específicos como efeitos aleatórios, que são estimados e que descrevem todo o conjunto de espécies; para comparação de métodos, ajustamos uma versão de efeitos fixos do modelo. Para a determinação da idade das aves usamos o sistema WRP. Apresentamos uma nova variante do modelo CJS com um parâmetro de mistura para a sobrevivência de aves de idade incerta no momento da primeira captura. Encontramos efeito forte da idade na sobrevivência, com probabilidades de sobrevivência menor para os jovens do que para os adultos; evidência de efeito latitude sobre a sobrevivência, que suporta a hipótese amplamente aceita de variação na sobrevivência com a latitude; e discutimos diferenças metodológicas interessantes entre modelo de efeitos aleatórios e fixos relacionados com a precisão das estimativas e o âmbito de inferência, que nos levam a concluir que os modelos de efeitos aleatórios são os mais adequados para a nossa análise. Concluímos que não é necessário invocar uma hipótese alternativa de maior sobrevivência juvenil nos trópicos a fim de explicar o gradiente latitudinal no tamanho da ninhada. / The observation of latitudinal gradients in bird life history traits has motivated the study of avian life history evolution and variability. A well-documented example is the variation in clutch size, where lower latitude birds tend to have smaller clutches than their higher latitude counterparts. A hypothesis that explains this variation proposes that survival in tropical latitudes is higher to compensate for smaller clutch size and prevent population extinctions. This explanation has had a wide acceptance and support by some studies, but has been questioned by others who have not found such higher survival rates in tropical birds. In an implicit manner, all these studies have based their results on adult survival. Populations with smaller clutch size would not be able to grow as well as populations with larger clutches; therefore one is justified to believe that something else must change with latitude. In the search for alternative explanations to the persistence of tropical bird populations with relatively small clutch sizes it has also been proposed that, if there were no differences in adult survival among latitudes, the fundamental trait that varies is juvenile survival, with higher survival rates for tropical juveniles birds than for temperate ones. However, currently there is little evidence that supports this conclusion. The contrasting results of those studies suggest a lack of a general consensus about the hypothesis that tropical birds have higher survival rates than birds of temperate regions, motivating the formulation of alternative hypotheses, and inviting further tests of the hypothesis. In our study we aim to a) assess the effect of age on survival in a tropical bird community, estimating age-specific annual apparent survival probabilities for a set of passerine understory birds from the central Brazilian Amazon; and b) contribute to the debate about the latitudinal gradient in adult survival by comparing our adult survival estimates to estimates of temperate-zone adult survival probabilities. To estimate the age-specific survival we fit to our data a hierarchical multispecies Cormack-Jolly-Seber (CJS) model for n species, that treats species-specific parameters as random effects that are estimated and that describe the whole assemblage of species; for comparison of methods, we also fit a fixed-effects version of the model. To age birds we use the cycle-based WRP system. We introduce a novel variant of CJS model with a mixture component for the survival of birds of uncertain age at the time of banding. We found strong effect of age on survival, with juveniles surviving less than adults; evidence of latitude effect on survival, that supports the widely accepted hypothesis of variation on survival with latitude; and methodological differences between random and fixed effects model related to precision of estimates and scope of inference, that lead us to conclude that random-effects models are more appropriate for our analysis. We conclude that there is no reason for an alternative latitudinal trend in juvenile survival to account for the general trend in clutch size.

Made available in DSpace on 2017-05-12T14:47:01Z (GMT). No. of bitstreams: 1 Sergio_ Bazilio.pdf: 3134289 bytes, checksum: 8461a5a239e9c7674bb3db224824dc63 (MD5) Previous issue date: 2015-02-12 / Benthic macroinvertebrate community is very important to streams, since they make part of the energy flow and are also an important food source for adjacent and upper trophic levels. They form a very diversified fauna and their structure may be influenced by several environmental factors, which vary in time, space and in an analyzed scale. Samples were taken from benthic macroinvertebrates and environmental descriptors during winter, spring and summer of 2012 and autumn of 2013 in ten sampling points in Lontra river watershed (Paraná, Brazil). Thus, this study aimed at: a) characterizing the watershed morphometry in Lontra river with data from the Shuttle Radar Topography Mission (SRTM); b) analyzing the structure of benthic macroinvertebrate community according to their taxonomic family level and functional groups; c) investigating seasonal variation (during the four seasons) and spatial variation of this community structure due to spatial differences in morphological and physiographic characteristics of the sampled sections; d) identifying the community variability of benthic macroinvertebrates at three spatial scales (river, river segment and mesohabitat) emphasizing spatial scales, which best explain the community structure in this watershed; e) investigating which measured environmental descriptors influence on the community structure and f) which variability percentage in organisms richness can be explained by the measured local environmental descriptors. / Os macroinvertebrados bentônicos constituem uma importante comunidade em riachos, pois participam do fluxo de energia, logo são um importante recurso alimentar para níveis tróficos adjacentes e superiores. Formam uma fauna bastante diversificada e a estrutura dessa comunidade pode ser influenciada por diversos fatores ambientais, os quais variam no tempo, no espaço e na escala analisada. Realizaram-se amostragens de macroinvertebrados bentônicos e descritores ambientais nos períodos de inverno, primavera e verão de 2012 e outono de 2013 em dez pontos amostrais na bacia do Rio Lontra (PR, Brasil). O presente estudo objetivou: a) caracterizar morfometricamente a microbacia do Rio Lontra com dados do projeto da Missão do Radar Transportado Espacial (SRTM). b) analisar a estrutura da comunidade de macroinvertebrados bentônicos em nível taxonômico de família e de grupos funcionais; c) investigar a variação sazonal (nas quatro estações do ano) e a variação espacial da estrutura da comunidade em função de diferenças espaciais nas características morfofisiográficas dos trechos amostrados; d) identificar a variabilidade da comunidade de macroinvertebrados bentônicos em três escalas espaciais (rio, segmento de rio e mesohabitat) com ênfase nas escalas espaciais que melhor explicam a estrutura da comunidade nesta bacia; e) investigar quais descritores ambientais mensurados influenciam a estrutura da comunidade e f) qual porcentagem da variabilidade na riqueza de organismos pode ser explicada pelos descritores ambientais locais mensurados.

L'esca de la vigne fait partie des maladies de dépérissement incurables dont l'étiologie n'est pas complément élucidée. Elle représente un des problèmes majeurs en viticulture. L'objectif général de cette thèse est d'améliorer la compréhension des processus épidémiques et des facteurs de risque. Pour ce faire, nous avons mené une étude quantitative du développement spatio-temporel de l'esca à l'échelle de la parcelle. Dans un premier temps, pour détecter d'éventuelles corrélations spatiales entre les cas de maladie, des tests statistiques non paramétriques sont appliqués aux données spatio-temporelles d'expression foliaires de l'esca pour 15 parcelles du bordelais. Une diversité de profils spatiaux, allant d'une distribution aléatoire à fortement structurée est trouvée. Dans le cas de structures très agrégées, les tests n'ont pas montré d'augmentation significative de la taille des foyers, ni de propagation secondaire locale à partir de ceps symptomatiques, suggérant un effet de l'environnement dans l'explication de cette agrégation. Dans le but de modéliser l'occurrence des symptômes foliaires, nous avons développé des modèles logistiques hiérarchiques intégrant à la fois des covariables exogènes liées à l'environnement et des covariables de voisinage de ceps déjà malades mais aussi un processus latent pour l'auto-corrélation spatio-temporelle. Les inférences bayésiennes sont réalisées en utilisant la méthode INLA (Inverse Nested Laplace Approximation). Les résultats permettent de conforter l'hypothèse du rôle significatif des facteurs environnementaux dans l'augmentation du risque d'occurrence des symptômes. L'effet de propagation de l'esca à petite échelle à partir de ceps déjà atteints situés sur le rang ou hors rang n'est pas montré. Un modèle autologistique de régression, deux fois centré, qui prend en compte de façon plus explicite la structure spatio-temporelle de voisinage, est également développé. Enfin, une méthode géostatistique d'interpolation de données de nature anisotropique atypique est proposée. Elle permet d'interpoler la variable auxiliaire de résistivité électrique du sol pour estimer à l'échelle de chaque plante de la parcelle, la réserve en eau du sol disponible pour la vigne. Les méthodes géostatistique et spatio-temporelles développées dans cette thèse ouvrent des perspectives pour identifier les facteurs de risques et prédire le développement de l'esca de la vigne dans des contextes agronomiques variés. / Esca grapevine disease is one of the incurable dieback disease with the etiology not completely elucidated. It represents one of the major threats for viticulture around the world. To better understand the underlying process of esca spread and the risk factors of this disease, we carried out quantitative analyses of the spatio-temporal development of esca at vineyard scale. In order to detect the spatial correlation among the diseased vines, the non-parametric statistical tests were applied to the spatio-temporal data of esca foliar symptom expression for 15 vineyards in Bordeaux region. Among vineyards, a large range of spatial patterns, from random to strongly structured, were found. In the vineyards with strongly aggregated patterns, no significant increase in the size of cluster and local spread from symptomatic vines was shown, suggesting an effect of the environment in the explanation of this aggregation. To model the foliar symptom occurrence, we developed hierarchical logistic regression models by integrating exogenous covariates, covariates of neighboring symptomatic vines already diseased, and also a latent process with spatio-temporal auto-correlation. The Bayesian inferences of these models were performed by INLA (Inverse Nested Laplace Approximation) approach. The results confirmed the effect of environmental factors on the occurrence risk of esca symptom. The secondary locally spread of esca from symptomatic vines located on the same row or out of row was not shown. A two-step centered auto-logistic regression model, which explicitly integrated the spatio-temporal neighboring structure, was also developed. At last, a geostatistical method was proposed to interpolate data with a particular anisotropic structure. It allowed interpolating the ancillary variable, electrical resistivity of soil, which were used to estimate the available soil water content at vine-scale. These geostatistical methods and spatio-temporal statistical methods developed in this thesis offered outlook to identify risk factors, and thereafter to predict the development of esca grapevine disease in different agronomical contexts.

Orientador: José Sílvio Govone / Resumo: As variáveis aleatórias no espaço estão definidas como funções aleatórias sujeitas à teoria das variáveis regionalizadas. Para assumir continuidade espacial com um número limitado de realizações da variável aleatória são necessárias as hipóteses de estacionariedade, as quais envolvem diferentes graus de homogeneidade espacial. Formalmente, uma variável regionalizada Z é estacionária se os momentos estatísticos de Z(s+h) forem os mesmos para qualquer vetor h. A hipótese de estacionariedade de primeira ordem é definida como a hipótese de que o momento de primeira ordem da distribuição da função aleatória Z(s) é constante em toda a área. A hipótese intrínseca é baseada no cálculo de médias globais das semivariancias, com a pressuposição de estacionariedade de 1a ordem e da estacionariedade da variância dos incrementos. Embora muitas variáveis sejam suscetível a dupla ou múltipla estacionariedade, estas estruturas espaciais não são levadas em consideração pelo semivariograma usual. Na perspectiva de solucionar o problema apontado, buscou-se identificar os locais dos pontos de mudança na média que definem mais de uma estrutura de semivariancia, com o objetivo de melhorar a qualidade dos mapas de Krigagem Ordinária. Para isso, foi utilizado o Método de Partição Produto (MPP), com enfoque espacial, denominado Método de Partição Produto Espacial (MPPs). Para separar os grupos, foi criada uma função de busca de ponto de mudança na média utilizando o modelo hierárquico bayesiano, denom... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: The random variables in space are defined by random functions subject to regionalized variable theory. To assume spatial continuity with a limited number of realization of the random variable, we need to assume stationarity hypotheses, which involve different degrees of spatial homogeneity. Formally, a regionalized variable Z is stationary if statistical moments of Z(s + h) are the same for any vector h. The first order stationarity hypothesis is defined to be the hypothesis that first order moment of the distribution of the random function Z(s) is constant throughout the area. The intrinsic hypothesis is based on the computation of global means of semivariate models, with the assumption of 1st order stationarity and incremental variation stationarity. Although many variables are capable of double or multiple stationarity, these spatial structures are not taken into account by the usual semivariogram, and, consequently, cause acuracy problems in Kriging maps. In order xvii to solve the described problem, it was identify the points of change in the average with the objective of improving the quality and accuracy of the maps of Ordinary Kriging. To separate the groups, a mean change point search function was created using the Bayesian hierarchical model, called the Space Product Partition Model (MPPs). Two databases were used to test the model’s potential to separate spatially dependent groups, in which the former suspected a change in mean while in the latter. “ Data2 ”, there... (Complete abstract click electronic access below) / Doutor

Bayesian statistics is an important approach to modern statistical analyses. It allows us to use our prior knowledge of the unknown parameters to construct a model for our data set. The foundation of Bayesian analysis is Bayes' Rule, which in its proportional form indicates that the posterior is proportional to the prior times the likelihood. We will demonstrate how we can apply Bayesian statistical techniques to fit a linear regression model and a hierarchical linear regression model to a data set. We will show how to apply different distributions to Bayesian analyses and how the use of a prior affects the model. We will also make a comparison between the Bayesian approach and the traditional frequentist approach to data analyses.

The ability to learn is essential for robots to function and perform services within a dynamic human environment. Robot programming by demonstration facilitates learning through a human teacher without the need to develop new code for each task that the robot performs. In order for learning to be generalizable, the robot needs to be able to grasp the underlying structure of the task being learned. This requires appropriate knowledge abstraction and representation. The goal of this thesis is to develop a learning by imitation system that abstracts knowledge of human demonstrations of a task and represents the abstracted knowledge in a hierarchical framework. The learning by imitation system is capable of performing both action and object recognition based on video stream data at the lower level of the hierarchy, while the sequence of actions and object states observed is reconstructed at the higher level of the hierarchy in order to form a coherent representation of the task. Furthermore, error recovery capabilities are included in the learning by imitation system to improve robustness to unexpected situations during task execution. The first part of the thesis focuses on motion learning to allow the robot to both recognize the actions for task representation at the higher level of the hierarchy and to perform the actions to imitate the task. In order to efficiently learn actions, the actions are segmented into meaningful atomic units called motion primitives. These motion primitives are then modeled using dynamic movement primitives (DMPs), a dynamical system model that can robustly generate motion trajectories to arbitrary goal positions while maintaining the overall shape of the demonstrated motion trajectory. The DMPs also contain weight parameters that are reflective of the shape of the motion trajectory. These weight parameters are clustered using affinity propagation (AP), an efficient exemplar clustering algorithm, in order to determine groups of similar motion primitives and thus, performing motion recognition. The approach of DMPs combined with APs was experimentally verified on two separate motion data sets for its ability to recognize and generate motion primitives. The second part of the thesis outlines how the task representation is created and used for imitating observed tasks. This includes object and object state recognition using simple computer vision techniques as well as the automatic construction of a Petri net (PN) model to describe an observed task. Tasks are composed of a sequence of actions that have specific pre-conditions, i.e. object states required before the action can be performed, and post-conditions, i.e. object states that result from the action. The PNs inherently encode pre-conditions and post-conditions of a particular event, i.e. action, and can model tasks as a coherent sequence of actions and object states. In addition, PNs are very flexible in modeling a variety of tasks including tasks that involve both sequential and parallel components. The automatic PN creation process has been tested on both a sequential two block stacking task and a three block stacking task involving both sequential and parallel components. The PN provides a meaningful representation of the observed tasks that can be used by a robot to imitate the tasks. Lastly, error recovery capabilities are added to the learning by imitation system in order to allow the robot to readjust the sequence of actions needed during task execution. The error recovery component is able to deal with two types of errors: unexpected, but known situations and unexpected, unknown situations. In the case of unexpected, but known situations, the learning system is able to search through the PN to identify the known situation and the actions needed to complete the task. This ability is useful not only for error recovery from known situations, but also for human robot collaboration, where the human unexpectedly helps to complete part of the task. In the case of situations that are both unexpected and unknown, the robot will prompt the human demonstrator to teach how to recover from the error to a known state. By observing the error recovery procedure and automatically extending the PN with the error recovery information, the situation encountered becomes part of the known situations and the robot is able to autonomously recover from the error in the future. This error recovery approach was tested successfully on errors encountered during the three block stacking task.

Visual object category recognition is one of the most challenging problems in computer vision. Even assuming that we can obtain a near-perfect instance level representation with the advances in visual input devices and low-level vision techniques, object categorization still remains as a difficult problem because it requires drawing boundaries between instances in a continuous world, where the boundaries are solely defined by human conceptualization. Object categorization is essentially a perceptual process that takes place in a human-defined semantic space. In this semantic space, the categories reside not in isolation, but in relation to others. Some categories are similar, grouped, or co-occur, and some are not. However, despite this semantic nature of object categorization, most of the today's automatic visual category recognition systems rely only on the category labels for training discriminative recognition with statistical machine learning techniques. In many cases, this could result in the recognition model being misled into learning incorrect associations between visual features and the semantic labels, from essentially overfitting to training set biases. This limits the model's prediction power when new test instances are given. Using semantic knowledge has great potential to benefit object category recognition. First, semantic knowledge could guide the training model to learn a correct association between visual features and the categories. Second, semantics provide much richer information beyond the membership information given by the labels, in the form of inter-category and category-attribute distances, relations, and structures. Finally, the semantic knowledge scales well as the relations between categories become larger with an increasing number of categories. My goal in this thesis is to learn discriminative models for categorization that leverage semantic knowledge for object recognition, with a special focus on the semantic relationships among different categories and concepts. To this end, I explore three semantic sources, namely attributes, taxonomies, and analogies, and I show how to incorporate them into the original discriminative model as a form of structural regularization. In particular, for each form of semantic knowledge I present a feature learning approach that defines a semantic embedding to support the object categorization task. The regularization penalizes the models that deviate from the known structures according to the semantic knowledge provided. The first semantic source I explore is attributes, which are human-describable semantic characteristics of an instance. While the existing work treated them as mid-level features which did not introduce new information, I focus on their potential as a means to better guide the learning of object categories, by enforcing the object category classifiers to share features with attribute classifiers, in a multitask feature learning framework. This approach essentially discovers the common low-dimensional features that support predictions in both semantic spaces. Then, I move on to the semantic taxonomy, which is another valuable source of semantic knowledge. The merging and splitting criteria for the categories on a taxonomy are human-defined, and I aim to exploit this implicit semantic knowledge. Specifically, I propose a tree of metrics (ToM) that learns metrics that capture granularity-specific similarities at different nodes of a given semantic taxonomy, and uses a regularizer to isolate granularity-specific disjoint features. This approach captures the intuition that the features used for the discrimination of the parent class should be different from the features used for the children classes. Such learned metrics can be used for hierarchical classification. The use of a single taxonomy can be limited in that its structure is not optimal for hierarchical classification, and there may exist no single optimal semantic taxonomy that perfectly aligns with visual distributions. Thus, I next propose a way to overcome this limitation by leveraging multiple taxonomies as semantic sources to exploit, and combine the acquired complementary information across multiple semantic views and granularities. This allows us, for example, to synthesize semantics from both 'Biological', and 'Appearance'-based taxonomies when learning the visual features. Finally, as a further exploration of more complex semantic relations different from the previous two pairwise similarity-based models, I exploit analogies, which encode the relational similarities between two related pairs of categories. Specifically, I use analogies to regularize a discriminatively learned semantic embedding space for categorization, such that the displacements between the two category embeddings in both category pairs of the analogy are enforced to be the same. Such a constraint allows for a more confusing pair of categories to benefit from a clear separation in the matched pair of categories that share the same relation. All of these methods are evaluated on challenging public datasets, and are shown to effectively improve the recognition accuracy over purely discriminative models, while also guiding the recognition to be more semantic to human perception. Further, the applications of the proposed methods are not limited to visual object categorization in computer vision, but they can be applied to any classification problems where there exists some domain knowledge about the relationships or structures between the classes. Possible applications of my methods outside the visual recognition domain include document classification in natural language processing, and gene-based animal or protein classification in computational biology. / text