Regularized model learning in EDAs for continuous and multi-objective optimization

Probabilistic modeling is the de�ning characteristic of estimation of distribution algorithms (EDAs) which determines their behavior and performance in optimization. Regularization is a well-known statistical technique used for obtaining an improved model by reducing the generalization error of estimation, especially in high-dimensional problems. `1-regularization is a type of this technique with the appealing variable selection property which results in sparse model estimations. In this thesis, we study the use of regularization techniques for model learning in EDAs. Several methods for regularized model estimation in continuous domains based on a Gaussian distribution assumption are presented, and analyzed from di�erent aspects when used for optimization in a high-dimensional setting, where the population size of EDA has a logarithmic scale with respect to the number of variables. The optimization results obtained for a number of continuous problems with an increasing number of variables show that the proposed EDA based on regularized model estimation performs a more robust optimization, and is able to achieve signi�cantly better results for larger dimensions than other Gaussian-based EDAs. We also propose a method for learning a marginally factorized Gaussian Markov random �eld model using regularization techniques and a clustering algorithm. The experimental results show notable optimization performance on continuous additively decomposable problems when using this model estimation method. Our study also covers multi-objective optimization and we propose joint probabilistic modeling of variables and objectives in EDAs based on Bayesian networks, speci�cally models inspired from multi-dimensional Bayesian network classi�ers. It is shown that with this approach to modeling, two new types of relationships are encoded in the estimated models in addition to the variable relationships captured in other EDAs: objectivevariable and objective-objective relationships. An extensive experimental study shows the e�ectiveness of this approach for multi- and many-objective optimization. With the proposed joint variable-objective modeling, in addition to the Pareto set approximation, the algorithm is also able to obtain an estimation of the multi-objective problem structure. Finally, the study of multi-objective optimization based on joint probabilistic modeling is extended to noisy domains, where the noise in objective values is represented by intervals. A new version of the Pareto dominance relation for ordering the solutions in these problems, namely �-degree Pareto dominance, is introduced and its properties are analyzed. We show that the ranking methods based on this dominance relation can result in competitive performance of EDAs with respect to the quality of the approximated Pareto sets. This dominance relation is then used together with a method for joint probabilistic modeling based on `1-regularization for multi-objective feature subset selection in classi�cation, where six di�erent measures of accuracy are considered as objectives with interval values. The individual assessment of the proposed joint probabilistic modeling and solution ranking methods on datasets with small-medium dimensionality, when using two di�erent Bayesian classi�ers, shows that comparable or better Pareto sets of feature subsets are approximated in comparison to standard methods.

Análise de jatos de baixos níveis associados aum sistema convectivo de mesoescala na américa do sul: um estudo de caso

The Levei Low Jet (LLJ) observed in the Porto Alegre metropolitan region, Rio Grande do Sul State, Brazil, was analyzed using 1989-2003 at 00:00 and 12:00 UTC upper-air observations. The LLJ classification criteria proposed by Bonner (1968) and modified by Whiteman et aI. (1997) were applied to determine the LLJ occurrence. Afterwards was selected a LLJ event, that was one of the most intense observed in the summer (01/27/2002 at 12:00 UTC), during the study period. ln this study were used as tools: atmospheric soundings, GOES-8 satellite images, and wind, temperature and specific humidity fields from GLOBAL, ETA and BRAMS models. Based on the numerical analysis was possible to verify that the three models overestimated the specific humidity and potential temperature values, at LLJ time occurrence. The wind speed was underestimated by the models. It was observed in the study region, at 12:00 UTC (LLJ detected hour in the Porto Alegre region), by three models, warm and wet air from north, generating conditions to Mesoscale Convective System (MCS) formation and intensification.

Effect of Multiple Cranial Burr Hole Surgery on Prevention of Recurrent Ischemic Attacks in Children with Moyamoya Disease

Moyamoya disease (MMD) is an uncommon cerebrovascular disorder characterized by progressive stenosis of the terminal portion of the internal carotid artery and its main branches. Direct and indirect bypass techniques have been devised with the aim of promoting neoangiogenesis. The current study aimed to investigate the role of multiple cranial burr hole (MCBH) operations in the prevention of cerebral ischemic attacks in children with MMD. Seven children suffering from progressive MMD were submitted to the MCBH and arachnoid opening technique. Ten to 20 burr holes were drilled in the fronto-temporo-parieto-occipital area of each hemisphere in each patient, depending on the site and extent of the disease. All patients were evaluated pre- and postoperatively by means of Barthel index (BI), CT, MR, angio-MR, and angiography. Patients had no recurrence of ischemic attacks postoperatively. Neoangiogenesis was observed in both hemispheres. One patient developed a persistent subdural collection after surgery, thus requiring placement of a subdural-peritoneal shunt. Postoperative BI was statistically significantly improved (P = 0.02). This report suggests that MCBH for revascularization in MMD is a simple procedure with a relatively low risk of complications and effective for preventing cerebral ischemic attacks in children. In addition, MCBH may be placed as an adjunct to other treatments for MMD.

Análisis sobre los obstáculos para la construcción de una política pública penitenciaria en Colombia (1998-2014)

Desde 1998 la Corte Constitucional ha declarado en dos ocasiones “el estado de las cosas inconstitucional” ante las precarias condiciones del Sistema Nacional Penitenciario y Carcelario (SNPC), sin embargo, los esfuerzos institucionales por superar dicho estado han tenido efectos nulos o limitados. Prueba de ello son las altas tasas de hacinamiento y reincidencia que siguen manifestándose crónicamente por el deficiente funcionamiento del sistema. Precisamente este diagnóstico con alternativas de solución presume que esta situación se debe a la ausencia de una política pública integral, al partir de la identificación de los principales obstáculos para la construcción de una política pública penitenciaria en Colombia entre los años de 1998 y 2014. El ejercicio antes mencionado se apoya en la utilización de dos herramientas metodológicas a saber: el análisis estructural “MICMAC” y el análisis de involucrados. De los resultados arrojados por estos métodos se elaboran así mismo tres recomendaciones sobre política pública penitenciaria.

II Jornadas de Cooperaci??n Educativa sobre Mujeres, Paz y Educaci??n

Las Jornadas contaron con la participaci??n de representantes de los Ministerios y Secretar??as de Educaci??n y de los Organismos de Igualdad de Bolivia, Brasil, Chile, Colombia, Costa Rica, Cuba, Ecuador, El Salvador, Espa??a, Guatemala, Honduras, M??xico, Nicaragua, Panam??, Paraguay, Rep??blica Dominicana, Uruguay y Venezuela. En las distintas ponencias y talleres se exploran las posibles acciones que por parte de los Ministerios se realizan dentro de los sistemas educativos iberoamericanos, para colaborar eficiente y eficazmente con las tareas de los organismos o mecanismos gubernamentales de igualdad en la prevenci??n y atenci??n de las formas diversas de la violencia de g??nero y domestica.

A review of estimation of distribution algorithms in bioinformatics

Evolutionary search algorithms have become an essential asset in the algorithmic toolbox for solving high-dimensional optimization problems in across a broad range of bioinformatics problems. Genetic algorithms, the most well-known and representative evolutionary search technique, have been the subject of the major part of such applications. Estimation of distribution algorithms (EDAs) offer a novel evolutionary paradigm that constitutes a natural and attractive alternative to genetic algorithms. They make use of a probabilistic model, learnt from the promising solutions, to guide the search process. In this paper, we set out a basic taxonomy of EDA techniques, underlining the nature and complexity of the probabilistic model of each EDA variant. We review a set of innovative works that make use of EDA techniques to solve challenging bioinformatics problems, emphasizing the EDA paradigm's potential for further research in this domain.

Multi-objective Estimation of Distribution Algorithm Based on Joint Modeling of Objectives and Variables

This paper proposes a new multi-objective estimation of distribution algorithm (EDA) based on joint modeling of objectives and variables. This EDA uses the multi-dimensional Bayesian network as its probabilistic model. In this way it can capture the dependencies between objectives, variables and objectives, as well as the dependencies learnt between variables in other Bayesian network-based EDAs. This model leads to a problem decomposition that helps the proposed algorithm to find better trade-off solutions to the multi-objective problem. In addition to Pareto set approximation, the algorithm is also able to estimate the structure of the multi-objective problem. To apply the algorithm to many-objective problems, the algorithm includes four different ranking methods proposed in the literature for this purpose. The algorithm is applied to the set of walking fish group (WFG) problems, and its optimization performance is compared with an evolutionary algorithm and another multi-objective EDA. The experimental results show that the proposed algorithm performs significantly better on many of the problems and for different objective space dimensions, and achieves comparable results on some compared with the other algorithms.

Edge Detection by Adaptive Splitting II. The Three-Dimensional Case

In Llanas and Lantarón, J. Sci. Comput. 46, 485–518 (2011) we proposed an algorithm (EDAS-d) to approximate the jump discontinuity set of functions defined on subsets of ℝ d . This procedure is based on adaptive splitting of the domain of the function guided by the value of an average integral. The above study was limited to the 1D and 2D versions of the algorithm. In this paper we address the three-dimensional problem. We prove an integral inequality (in the case d=3) which constitutes the basis of EDAS-3. We have performed detailed computational experiments demonstrating effective edge detection in 3D function models with different interface topologies. EDAS-1 and EDAS-2 appealing properties are extensible to the 3D case

La protección radiológica en la rehabilitación de Fukushima

La protección radiológica en la rehabilitación de Fukushima

Hybrid sampling on mutual information entropy-based clustering ensembles for optimizations

In this paper, we focus on the design of bivariate EDAs for discrete optimization problems and propose a new approach named HSMIEC. While the current EDAs require much time in the statistical learning process as the relationships among the variables are too complicated, we employ the Selfish gene theory (SG) in this approach, as well as a Mutual Information and Entropy based Cluster (MIEC) model is also set to optimize the probability distribution of the virtual population. This model uses a hybrid sampling method by considering both the clustering accuracy and clustering diversity and an incremental learning and resample scheme is also set to optimize the parameters of the correlations of the variables. Compared with several benchmark problems, our experimental results demonstrate that HSMIEC often performs better than some other EDAs, such as BMDA, COMIT, MIMIC and ECGA. © 2009 Elsevier B.V. All rights reserved.

Characterization and Technological Features of Autochthonous Coagulase-Negative Staphylococci as Potential Starters for Portuguese Dry Fermented Sausages

The manufacture of dry fermented sausages is an important part of the meat industry in Southern Europeancountries. These products are usually produced in small shops from a mixture of pork, fat, salt, and condiments andare stuffed into natural casings. Meat sausages are slowly cured through spontaneous fermentation by autochthonousmicrobiota present in the raw materials or introduced during manufacturing. The aim of this work was to evaluate thetechnological and safety features of coagulase-negative staphylococci (CNS) isolated from Portuguese dry fermented meatsausages in order to select autochthonous starters. Isolates (n = 104) obtained from 2 small manufacturers were identifiedas Staphylococcus xylosus, Staphylococcus equorum, Staphylococcus saprophyticus,andStaphylococcus carnosus. Genomically diverseisolates (n = 82) were selected for further analysis to determine the ability to produce enzymes (for example, nitrate-reductases, proteases, lipases) and antibiotic susceptibility. Autochthonous CNS producing a wide range of enzymes andshowing low antibioresistance were selected as potential starters for future use in the production of dry fermented meatsausages.

Ecosistemas dependientes de aguas subterráneas

Tiene habido muchas definiciones para el concepto de Ecosistemas Dependientes de Aguas Subterráneas (EDAS, GDE en inglés), pero resumiendo son ecosistemas que usan agua subterránea en alguna parte de su ciclo de vida o por toda una generación e donde esta es crítica para la existencia de esas especies. El uso del agua subterránea no equivale necesariamente a una dependencia de las aguas subterráneas (Colvin et al. 2003). Por dependencia se entiende que el ecosistema sería significativamente alterado o mismo irreversiblemente degradado si la disponibilidad o calidad del agua subterránea fuera alterada más allá de su rango "normal" de fluctuación, o sea, son ecosistemas que dependen en el todo o en parte de las aguas subterráneas para mantener un nivel adecuado de la función del ecosistema y el mantenimiento de la composición de la comunidad (Smith et al. 2006). La dependencia de los EDAS de las aguas subterráneas es muy variable, oscilando entre parcial y con poca frecuencia a continua y totalmente dependiente. Estos ecosistemas, incluyendo los humedales, vegetación en general, vegetación de manantiales, flujos de base de los ríos, ecosistemas de acuíferos y cuevas, vertidos salinos de lagunas costeras, manantiales, manglares, charcos en ríos, lagos en herradura y pantanos colgados (Sinclair Knight Merz 2001) y descargas de agua subterránea en el océano, representan componentes complejas e importantes de la diversidad biológica. Una de las clasificaciones posibles para los EDAS, sería considerar los sistemas terrestres, sistemas acuíferos y de cuevas, sistemas rivereños y lagunares interiores (incluyendo humedales y pantanos), sistemas costeros (lagunas y estuarios) y los sistemas marinos. Posibles amenazas a los EDAS incluyen la extracción y la contaminación química y con nutrientes del agua subterránea, la salinización, la alteración de la gestión de las aguas superficiales y subterráneas, las alteraciones climáticas, lo que puede afectar una cadena complexa de interacciones en el mundo natural.