Bilinear models for spatio-temporal point distribution analysis: application to extrapolation of left ventricular, biventricular and whole heart cardiac dynamics

In this work we describe the usage of bilinear statistical models as a means of factoring the shape variability into two components attributed to inter-subject variation and to the intrinsic dynamics of the human heart. We show that it is feasible to reconstruct the shape of the heart at discrete points in the cardiac cycle. Provided we are given a small number of shape instances representing the same heart atdifferent points in the same cycle, we can use the bilinearmodel to establish this. Using a temporal and a spatial alignment step in the preprocessing of the shapes, around half of the reconstruction errors were on the order of the axial image resolution of 2 mm, and over 90% was within 3.5 mm. From this, weconclude that the dynamics were indeed separated from theinter-subject variability in our dataset.

Estudo e implementação de um controlador preditivo generalizado bilinear compensado adaptativo

The present work presents the study and implementation of an adaptive bilinear compensated generalized predictive controller. This work uses conventional techniques of predictive control and includes techniques of adaptive control for better results. In order to solve control problems frequently found in the chemical industry, bilinear models are considered to represent the dynamics of the studied systems. Bilinear models are simpler than general nonlinear model, however it can to represent the intrinsic not-linearities of industrial processes. The linearization of the model, by the approach to time step quasilinear , is used to allow the application of the equations of the generalized predictive controller (GPC). Such linearization, however, generates an error of prediction, which is minimized through a compensation term. The term in study is implemented in an adaptive form, due to the nonlinear relationship between the input signal and the prediction error.Simulation results show the efficiency of adaptive predictive bilinear controller in comparison with the conventional.

Improving sampling, optimization and feature extraction in Boltzmann machines

L’apprentissage supervisé de réseaux hiérarchiques à grande échelle connaît présentement un succès fulgurant. Malgré cette effervescence, l’apprentissage non-supervisé représente toujours, selon plusieurs chercheurs, un élément clé de l’Intelligence Artificielle, où les agents doivent apprendre à partir d’un nombre potentiellement limité de données. Cette thèse s’inscrit dans cette pensée et aborde divers sujets de recherche liés au problème d’estimation de densité par l’entremise des machines de Boltzmann (BM), modèles graphiques probabilistes au coeur de l’apprentissage profond. Nos contributions touchent les domaines de l’échantillonnage, l’estimation de fonctions de partition, l’optimisation ainsi que l’apprentissage de représentations invariantes. Cette thèse débute par l’exposition d’un nouvel algorithme d'échantillonnage adaptatif, qui ajuste (de fa ̧con automatique) la température des chaînes de Markov sous simulation, afin de maintenir une vitesse de convergence élevée tout au long de l’apprentissage. Lorsqu’utilisé dans le contexte de l’apprentissage par maximum de vraisemblance stochastique (SML), notre algorithme engendre une robustesse accrue face à la sélection du taux d’apprentissage, ainsi qu’une meilleure vitesse de convergence. Nos résultats sont présent ́es dans le domaine des BMs, mais la méthode est générale et applicable à l’apprentissage de tout modèle probabiliste exploitant l’échantillonnage par chaînes de Markov. Tandis que le gradient du maximum de vraisemblance peut-être approximé par échantillonnage, l’évaluation de la log-vraisemblance nécessite un estimé de la fonction de partition. Contrairement aux approches traditionnelles qui considèrent un modèle donné comme une boîte noire, nous proposons plutôt d’exploiter la dynamique de l’apprentissage en estimant les changements successifs de log-partition encourus à chaque mise à jour des paramètres. Le problème d’estimation est reformulé comme un problème d’inférence similaire au filtre de Kalman, mais sur un graphe bi-dimensionnel, où les dimensions correspondent aux axes du temps et au paramètre de température. Sur le thème de l’optimisation, nous présentons également un algorithme permettant d’appliquer, de manière efficace, le gradient naturel à des machines de Boltzmann comportant des milliers d’unités. Jusqu’à présent, son adoption était limitée par son haut coût computationel ainsi que sa demande en mémoire. Notre algorithme, Metric-Free Natural Gradient (MFNG), permet d’éviter le calcul explicite de la matrice d’information de Fisher (et son inverse) en exploitant un solveur linéaire combiné à un produit matrice-vecteur efficace. L’algorithme est prometteur: en terme du nombre d’évaluations de fonctions, MFNG converge plus rapidement que SML. Son implémentation demeure malheureusement inefficace en temps de calcul. Ces travaux explorent également les mécanismes sous-jacents à l’apprentissage de représentations invariantes. À cette fin, nous utilisons la famille de machines de Boltzmann restreintes “spike & slab” (ssRBM), que nous modifions afin de pouvoir modéliser des distributions binaires et parcimonieuses. Les variables latentes binaires de la ssRBM peuvent être rendues invariantes à un sous-espace vectoriel, en associant à chacune d’elles, un vecteur de variables latentes continues (dénommées “slabs”). Ceci se traduit par une invariance accrue au niveau de la représentation et un meilleur taux de classification lorsque peu de données étiquetées sont disponibles. Nous terminons cette thèse sur un sujet ambitieux: l’apprentissage de représentations pouvant séparer les facteurs de variations présents dans le signal d’entrée. Nous proposons une solution à base de ssRBM bilinéaire (avec deux groupes de facteurs latents) et formulons le problème comme l’un de “pooling” dans des sous-espaces vectoriels complémentaires.

Análise e desenvolvimento de controladores preditivos multivariáveis baseados em multi-modelos bilineares

This work addresses issues related to analysis and development of multivariable predictive controllers based on bilinear multi-models. Linear Generalized Predictive Control (GPC) monovariable and multivariable is shown, and highlighted its properties, key features and applications in industry. Bilinear GPC, the basis for the development of this thesis, is presented by the time-step quasilinearization approach. Some results are presented using this controller in order to show its best performance when compared to linear GPC, since the bilinear models represent better the dynamics of certain processes. Time-step quasilinearization, due to the fact that it is an approximation, causes a prediction error, which limits the performance of this controller when prediction horizon increases. Due to its prediction error, Bilinear GPC with iterative compensation is shown in order to minimize this error, seeking a better performance than the classic Bilinear GPC. Results of iterative compensation algorithm are shown. The use of multi-model is discussed in this thesis, in order to correct the deficiency of controllers based on single model, when they are applied in cases with large operation ranges. Methods of measuring the distance between models, also called metrics, are the main contribution of this thesis. Several application results in simulated distillation columns, which are close enough to actual behaviour of them, are made, and the results have shown satisfactory

Desenvolvimento e avaliação de controladores preditivos baseados em modelos biliares

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

Probing neutrino oscillations in supersymmetric models at the Large Hadron Collider

The lightest supersymmetric particle may decay with branching ratios that correlate with neutrino oscillation parameters. In this case the CERN Large Hadron Collider (LHC) has the potential to probe the atmospheric neutrino mixing angle with sensitivity competitive to its low-energy determination by underground experiments. Under realistic detection assumptions, we identify the necessary conditions for the experiments at CERN's LHC to probe the simplest scenario for neutrino masses induced by minimal supergravity with bilinear R parity violation.

CP properties of symmetry-constrained two-Higgs-doublet models

The two-Higgs-doublet model can be constrained by imposing Higgs-family symmetries and/or generalized CP symmetries. It is known that there are only six independent classes of such symmetry-constrained models. We study the CP properties of all cases in the bilinear formalism. An exact symmetry implies CP conservation. We show that soft breaking of the symmetry can lead to spontaneous CP violation (CPV) in three of the classes.

Probing bilinear R-parity violating supergravity at the LHC

We study the collider phenomenology of bilinear R-parity violating supergravity, the simplest effective model for supersymmetric neutrino masses accounting for the current neutrino oscillation data. At the CERN Large Hadron Collider the center-of-mass energy will be high enough to probe directly these models through the search for the superpartners of the Standard Model (SM) particles. We analyze the impact of R-parity violation on the canonical supersymmetry searches-that is, we examine how the decay of the lightest supersymmetric particle (LSP) via bilinear R-parity violating interactions degrades the average expected missing momentum of the reactions and show how this diminishes the reach in the usual channels for supersymmetry searches. However, the R-parity violating interactions lead to an enhancement of the final states containing isolated same-sign di-leptons and trileptons, compensating the reach loss in the fully inclusive channel. We show how the searches for displaced vertices associated to LSP decay substantially increase the coverage in supergravity parameter space, giving the corresponding reaches for two reference luminosities of 10 and 100 fb(-1) and compare with those of the R-parity conserving minimal supergravity model.

Probing bilinear R-parity violating supergravity at the LHC

We study the collider phenomenology of bilinear R-parity violating supergravity, the simplest effective model for supersymmetric neutrino masses accounting for the current neutrino oscillation data. At the CERN Large Hadron Collider the center-of-mass energy will be high enough to probe directly these models through the search for the superpartners of the Standard Model (SM) particles. We analyze the impact of R-parity violation on the canonical supersymmetry searches-that is, we examine how the decay of the lightest supersymmetric particle (LSP) via bilinear R-parity violating interactions degrades the average expected missing momentum of the reactions and show how this diminishes the reach in the usual channels for supersymmetry searches. However, the R-parity violating interactions lead to an enhancement of the final states containing isolated same-sign di-leptons and trileptons, compensating the reach loss in the fully inclusive channel. We show how the searches for displaced vertices associated to LSP decay substantially increase the coverage in supergravity parameter space, giving the corresponding reaches for two reference luminosities of 10 and 100 fb(-1) and compare with those of the R-parity conserving minimal supergravity model.

Probing neutrino oscillations in supersymmetric models at the Large Hadron Collider

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Probing neutralino properties in minimal supergravity with bilinear R-parity violation

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Confinement, solitons and the equivalence between the sine-Gordon and massive Thirring models

We consider a two-dimensional integrable and conformally invariant field theory possessing two Dirac spinors and three scalar fields. The interaction couples bilinear terms in the spinors to exponentials of the scalars. Its integrability properties are based on the sl(2) affine Kac-Moody algebra, and it is a simple example of the so-called conformal affine Toda theories coupled to matter fields. We show, using bosonization techniques, that the classical equivalence between a U(1) Noether current and the topological current holds true at the quantum level, and then leads to a bag model like mechanism for the confinement of the spinor fields inside the solitons. By bosonizing the spinors we show that the theory decouples into a sine-Gordon model and free scalars. We construct the two-soliton solutions and show that their interactions lead to the same time delays as those for the sine-Gordon solitons. The model provides a good laboratory to test duality ideas in the context of the equivalence between the sine-Gordon and Thirring theories. © 2000 Elsevier Science B.V. All rights reserved.

On Switched Regulator Design of Uncertain Nonlinear Systems Using Takagi-Sugeno Fuzzy Models

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Probing neutralino properties in minimal supergravity with bilinear R-parity violation

Supersymmetric models with bilinear R-parity violation can account for the observed neutrino masses and mixing parameters indicated by neutrino oscillation data. We consider minimal supergravity versions of bilinear R-parity violation where the lightest supersymmetric particle is a neutralino. This is unstable, with a large enough decay length to be detected at the CERN Large Hadron Collider. We analyze the Large Hadron Collider potential to determine the lightest supersymmetric particle properties, such as mass, lifetime and branching ratios, and discuss their relation to neutrino properties.

Integration of computational models and experimental characterization to study internal frost damage in cementitious materials

The objective of this doctoral research is to investigate the internal frost damage due to crystallization pore pressure in porous cement-based materials by developing computational and experimental characterization tools. As an essential component of the U.S. infrastructure system, the durability of concrete has significant impact on maintenance costs. In cold climates, freeze-thaw damage is a major issue affecting the durability of concrete. The deleterious effects of the freeze-thaw cycle depend on the microscale characteristics of concrete such as the pore sizes and the pore distribution, as well as the environmental conditions. Recent theories attribute internal frost damage of concrete is caused by crystallization pore pressure in the cold environment. The pore structures have significant impact on freeze-thaw durability of cement/concrete samples. The scanning electron microscope (SEM) and transmission X-ray microscopy (TXM) techniques were applied to characterize freeze-thaw damage within pore structure. In the microscale pore system, the crystallization pressures at sub-cooling temperatures were calculated using interface energy balance with thermodynamic analysis. The multi-phase Extended Finite Element Modeling (XFEM) and bilinear Cohesive Zone Modeling (CZM) were developed to simulate the internal frost damage of heterogeneous cement-based material samples. The fracture simulation with these two techniques were validated by comparing the predicted fracture behavior with the captured damage from compact tension (CT) and single-edge notched beam (SEB) bending tests. The study applied the developed computational tools to simulate the internal frost damage caused by ice crystallization with the two dimensional (2-D) SEM and three dimensional (3-D) reconstructed SEM and TXM digital samples. The pore pressure calculated from thermodynamic analysis was input for model simulation. The 2-D and 3-D bilinear CZM predicted the crack initiation and propagation within cement paste microstructure. The favorably predicted crack paths in concrete/cement samples indicate the developed bilinear CZM techniques have the ability to capture crack nucleation and propagation in cement-based material samples with multiphase and associated interface. By comparing the computational prediction with the actual damaged samples, it also indicates that the ice crystallization pressure is the main mechanism for the internal frost damage in cementitious materials.