This paper describes attempts to model the modules of early vision in terms of minimizing energy functions, in particular energy functions allowing discontinuities in the solution. It examines the success of using Hopfield-style analog networks for solving such problems. Finally it discusses the limitations of the energy function approach.

No description available.

The structure, potential energy functions and vibrational spectra of several cyclic and bicyclic molecules have been investigated using several spectroscopic techniques and high-level ab initio and density functional theory (DFT) calculations. Laser induced fluorescence and Raman spectroscopies were used to study the conformation of 2- indanol in the electronic ground and excited states. These, along with detailed ab initio calculations, confirmed the existence of four different stable conformations with the one undergoing an intermolecular hydrogen bonding being the most stable. A theoretical two-dimensional surface in terms of the ring-puckering and the hydroxyl group internal rotation vibrations was constructed. This work was extended to obtain preliminary insights on the conformations and ring-puckering frequencies of 3-cyclopenten-1-ol using ab initio and DFT calculations. Infrared and Raman spectra were also utilized to study the structures and vibrational spectra of -crotonolactone and 2,3-cyclopentenopyridine (pyrindan). Ab initio results showed that -crotonolactone is rigidly planar in the electronic ground state and has a nearly harmonic ring-puckering potential function. The calculated vibrational levels were shown to be in very good agreement with the experimental ring-puckering frequency from vapor-phase Raman observations. The structures, vibrational spectra, and potential energy functions of several cyclic molecules were reinvestigated using high-level ab initio computations, and detailed vibrational analyses based on DFT-B3LYP calculated frequencies were also carried out. A number of new insights were presented by re-evaluating the available experimental data for several cyclopentenes, silacyclobutanes and silacyclopentenes. It was found that the vibrational spectra of some deuterated cyclopentenes possess extensive coupling between several ring modes and other low-frequency modes. Reassignments of these spectra have been proposed. Frequencies from DFT-B3LYP calculations showed very good agreement with the experimental values for silacyclobutane and its derivatives. The presence of silicon and halogen atoms did not affect the accuracy of the DFT calculations. In addition, the ring-puckering potential energy function for silacyclopent-2-ene was studied and alternative assignments of the far-infrared results were proposed. The new assignments are in good agreement with computational results. Silacyclopent-2-ene and its -1,1-d2 isotopomer were shown to be slightly puckered with barriers of less than 50 cm-1.

Os métodos diretos são adequados à análise de estabilidade transitória em sistemas de potência, já que não requerem a resolução, integração numérica, do conjunto de equações diferenciais que representam o sistema. Os métodos diretos utilizam as idéias de Lyapunov associadas ao princípio de invariância de LaSalle para estimar a área de atração dos sistemas de potência. A grande dificuldade dos métodos diretos está em encontrar uma função auxiliar V, denominada função de Lyapunov que satisfaça as condições estabelecidas pelo Teorema de Lyapunov. Neste trabalho é realizada uma revisão bibliográfica das funções de Lyapunov utilizadas para análise de estabilidade transitória em sistemas de potência. Analisa-se o problema da existência de funções de Lyapunov quando as condutâncias de transferência são consideradas. Utilizando-se de uma extensão do princípio de Invariância de LaSalle, apresenta-se uma nova função a qual é uma função de Lyapunov no sentido mais geral da extensão do princípio de invariância de LaSalle quando as condutâncias de transferência da matriz admitância da rede reduzida são consideradas. Estudou-se também a existência de funções de Lyapunov no sentido mais geral de extensão do princípio de invariância de LaSalle para modelos que preserva a estrutura da rede. Neste caso, infelizmente não encontramos uma função satisfazendo todas as hipóteses requeridas. / The direct methods are well-suited for transient stability analysis to power systems, since they do not require the solution of the set of differential equations of the system model. The direct methods use the Lyapunov\'s ideas related to the LaSalle\'s invariance principle to estimate the power system attraction area. The great difficulty of the direct methods is to find an auxiliar function V, called Lyapunov function, which satisfies the conditions of Lyapunov\'s theorem. In this work, a bibliographic review of the Lyapunov functions used in transient stability analysis of power systems is done. The problem of existence of Lyapunov functions, when the transfer conductances are considered, is analysed. Using LaSalle\'s invariance principle extension, a Lyapunov function considering the transfer conductances is presented. The existence of Lyapunov functions for models that preserv the network structure was studied using the LaSalle\'s invariance principle. Unfortunately, in these cases, we did not find a function satisfing all the required hypothesis.

Proteins are the fundamental machinery that enables the functions of life. It is critical to understand them not just for basic biology, but also to enable medical advances. The field of protein structure prediction is concerned with developing computational techniques to predict protein structure and function from a protein’s amino acid sequence, encoded for directly in DNA, alone. Despite much progress since the first computational models in the late 1960’s, techniques for the prediction of protein structure still cannot reliably produce structures of high enough accuracy to enable desired applications such as rational drug design. Protein structure refinement is the process of modifying a predicted model of a protein to bring it closer to its native state. In this dissertation a protein structure refinement technique, that of potential energy minimization using hybrid molecular mechanics/knowledge based potential energy functions is examined in detail. The generation of the knowledge-based component is critically analyzed, and in the end, a potential that is a modest improvement over the original is presented. This dissertation also examines the task of protein structure comparison. In evaluating various protein structure prediction techniques, it is crucial to be able to compare produced models against known structures to understand how well the technique performs. A novel technique is proposed that allows an in-depth yet intuitive evaluation of the local similarities between protein structures. Based on a graph analysis of pairwise atomic distance similarities, multiple regions of structural similarity can be identified between structures independently of relative orientation. Multidomain structures can be evaluated and this technique can be combined with global measures of similarity such as the global distance test. This method of comparison is expected to have broad applications in rational drug design, the evolutionary study of protein structures, and in the analysis of the protein structure prediction effort.

The scale of the energy business today and a favorable and promising economic environment for the production of natural gas, requires study of the thermophysical behavior of fluids: sophisticated experimentation yielding accurate, new volumetric data, and development and improvement of thermodynamic models. This work contains theoretical and experimental contributions in the form of 1) the revision and update of a field model to calculate compressibility factors starting from the gross heating value and the mole fractions of diluents in natural gas mixtures; 2) new reference quality volumetric data, gathered with state of the art techniques such as magnetic suspension densimetry and isochoric phase boundary determinations; 3) a rigorous first-principles uncertainty assessment for density measurements; and 4) a departure technique for the extension of these experimental data for calculating energy functions. These steps provide a complete experimental thermodynamic characterization of fluid samples. A modification of the SGERG model, a standard virial-type model for prediction of compressibility factors of natural gas mixtures, matches predictions from the master GERG-2008 equation of state, using least squares routines coded at NIST. The modification contains new values for parametric constants, such as molecular weights and the universal gas constant, as well as a new set of coefficients. A state-of-the-art high-pressure, single-sinker magnetic suspension densimeter is used to perform density measurements over a wide range of temperatures and pressures. This work contains data on nitrogen, carbon dioxide, and a typical residual gas mixture (95% methane, 4% ethane, and 1% propane). Experimental uncertainty results from a rigorous, first-principles estimation including composition uncertainty effects. Both low- and high-pressure isochoric apparatus are used to perform phase boundary measurements. Isochoric P-T data can determine the phase boundaries. Combined with density measurements, isochoric data provides isochoric densities. Further mathematical treatment, including noxious volume and thermal expansion corrections, and isothermal integration, leads to energy functions and thus to a full thermodynamic characterization.

Os métodos diretos são adequados à análise de estabilidade transitória em sistemas de potência, já que não requerem a resolução, integração numérica, do conjunto de equações diferenciais que representam o sistema. Os métodos diretos utilizam as idéias de Lyapunov associadas ao princípio de invariância de LaSalle para estimar a área de atração dos sistemas de potência. A grande dificuldade dos métodos diretos está em encontrar uma função auxiliar V, denominada função de Lyapunov que satisfaça as condições estabelecidas pelo Teorema de Lyapunov. Neste trabalho é realizada uma revisão bibliográfica das funções de Lyapunov utilizadas para análise de estabilidade transitória em sistemas de potência. Analisa-se o problema da existência de funções de Lyapunov quando as condutâncias de transferência são consideradas. Utilizando-se de uma extensão do princípio de Invariância de LaSalle, apresenta-se uma nova função a qual é uma função de Lyapunov no sentido mais geral da extensão do princípio de invariância de LaSalle quando as condutâncias de transferência da matriz admitância da rede reduzida são consideradas. Estudou-se também a existência de funções de Lyapunov no sentido mais geral de extensão do princípio de invariância de LaSalle para modelos que preserva a estrutura da rede. Neste caso, infelizmente não encontramos uma função satisfazendo todas as hipóteses requeridas. / The direct methods are well-suited for transient stability analysis to power systems, since they do not require the solution of the set of differential equations of the system model. The direct methods use the Lyapunov\'s ideas related to the LaSalle\'s invariance principle to estimate the power system attraction area. The great difficulty of the direct methods is to find an auxiliar function V, called Lyapunov function, which satisfies the conditions of Lyapunov\'s theorem. In this work, a bibliographic review of the Lyapunov functions used in transient stability analysis of power systems is done. The problem of existence of Lyapunov functions, when the transfer conductances are considered, is analysed. Using LaSalle\'s invariance principle extension, a Lyapunov function considering the transfer conductances is presented. The existence of Lyapunov functions for models that preserv the network structure was studied using the LaSalle\'s invariance principle. Unfortunately, in these cases, we did not find a function satisfing all the required hypothesis.

The fully quantum mechanical 'direct-potential-fit' (DPF) method has become increasingly widely used in the reduction of diatomic spectra. The central problem of this method is the representation of the potential energy and Born-Oppenheimer breakdown (BOB) correction functions. There are a number of problems associated with the existing method and potential forms. This thesis delineates these problems and finds solutions to some of them. In particular, it is shown that use of a different expansion variable and a new treatment of some of the expansions resolves most of the problems. These techniques have been successfully tested on the ground electronic states of the coinage metal hydrides and the Rb2 molecule. To address the problem of representing 'barrier' potential curves, a flexible new functional form, the 'double-exponential long-range' (DELR) potential function, is introduced and applied to the B barrier state of Li2. In addition, the Lambda-doubling level splitting which occurs for singlet Pi electronic states has been taken into account by extending the effective Schrodinger equation. The computer program DSPotFit developed in our laboratory for performing DPF analyses has been extended to incorporate the ability to fit the analytical potential energy functions to tunneling predissociation line widths for quasibound levels. Finally, an attempt is made to investigate whether there exists a hump in the ground state rotationless potential curve of beryllium hydride.

The fully quantum mechanical 'direct-potential-fit' (DPF) method has become increasingly widely used in the reduction of diatomic spectra. The central problem of this method is the representation of the potential energy and Born-Oppenheimer breakdown (BOB) correction functions. There are a number of problems associated with the existing method and potential forms. This thesis delineates these problems and finds solutions to some of them. In particular, it is shown that use of a different expansion variable and a new treatment of some of the expansions resolves most of the problems. These techniques have been successfully tested on the ground electronic states of the coinage metal hydrides and the Rb2 molecule. To address the problem of representing 'barrier' potential curves, a flexible new functional form, the 'double-exponential long-range' (DELR) potential function, is introduced and applied to the B barrier state of Li2. In addition, the Lambda-doubling level splitting which occurs for singlet Pi electronic states has been taken into account by extending the effective Schrodinger equation. The computer program DSPotFit developed in our laboratory for performing DPF analyses has been extended to incorporate the ability to fit the analytical potential energy functions to tunneling predissociation line widths for quasibound levels. Finally, an attempt is made to investigate whether there exists a hump in the ground state rotationless potential curve of beryllium hydride.

Internet offre une énorme quantité d’informations à portée de main et dans une telle variété de sujets, que tout le monde est en mesure d’accéder à une énorme variété de connaissances. Une telle grande quantité d’information pourrait apporter un saut en avant dans de nombreux domaines (moteurs de recherche, réponses aux questions, tâches NLP liées) si elle est bien utilisée. De cette façon, un enjeu crucial de la communauté d’intelligence artificielle a été de recueillir, d’organiser et de faire un usage intelligent de cette quantité croissante de connaissances disponibles. Heureusement, depuis un certain temps déjà des efforts importants ont été faits dans la collecte et l’organisation des connaissances, et beaucoup d’informations structurées peuvent être trouvées dans des dépôts appelés Bases des Connaissances (BCs). Freebase, Entity Graph Facebook ou Knowledge Graph de Google sont de bons exemples de BCs. Un grand problème des BCs c’est qu’ils sont loin d’êtres complets. Par exemple, dans Freebase seulement environ 30% des gens ont des informations sur leur nationalité. Cette thèse présente plusieurs méthodes pour ajouter de nouveaux liens entre les entités existantes de la BC basée sur l’apprentissage des représentations qui optimisent une fonction d’énergie définie. Ces modèles peuvent également être utilisés pour attribuer des probabilités à triples extraites du Web. On propose également une nouvelle application pour faire usage de cette information structurée pour générer des informations non structurées (spécifiquement des questions en langage naturel). On pense par rapport à ce problème comme un modèle de traduction automatique, où on n’a pas de langage correct comme entrée, mais un langage structuré. Nous adaptons le RNN codeur-décodeur à ces paramètres pour rendre possible cette traduction. / Internet provides a huge amount of information at hand in such a variety of topics, that now everyone is able to access to any kind of knowledge. Such a big quantity of information could bring a leap forward in many areas if used properly. This way, a crucial challenge of the Artificial Intelligence community has been to gather, organize and make intelligent use of this growing amount of available knowledge. Fortunately, important efforts have been made in gathering and organizing knowledge for some time now, and a lot of structured information can be found in repositories called Knowledge Bases (KBs). A main issue with KBs is that they are far from being complete. This thesis proposes several methods to add new links between the existing entities of the KB based on the learning of representations that optimize some defined energy function. We also propose a novel application to make use of this structured information to generate questions in natural language.