Estimating Dependency Structure as a Hidden Variable

This paper introduces a probability model, the mixture of trees that can account for sparse, dynamically changing dependence relationships. We present a family of efficient algorithms that use EMand the Minimum Spanning Tree algorithm to find the ML and MAP mixtureof trees for a variety of priors, including the Dirichlet and the MDL priors.

Notes on PCA, Regularization, Sparsity and Support Vector Machines

We derive a new representation for a function as a linear combination of local correlation kernels at optimal sparse locations and discuss its relation to PCA, regularization, sparsity principles and Support Vector Machines. We first review previous results for the approximation of a function from discrete data (Girosi, 1998) in the context of Vapnik"s feature space and dual representation (Vapnik, 1995). We apply them to show 1) that a standard regularization functional with a stabilizer defined in terms of the correlation function induces a regression function in the span of the feature space of classical Principal Components and 2) that there exist a dual representations of the regression function in terms of a regularization network with a kernel equal to a generalized correlation function. We then describe the main observation of the paper: the dual representation in terms of the correlation function can be sparsified using the Support Vector Machines (Vapnik, 1982) technique and this operation is equivalent to sparsify a large dictionary of basis functions adapted to the task, using a variation of Basis Pursuit De-Noising (Chen, Donoho and Saunders, 1995; see also related work by Donahue and Geiger, 1994; Olshausen and Field, 1995; Lewicki and Sejnowski, 1998). In addition to extending the close relations between regularization, Support Vector Machines and sparsity, our work also illuminates and formalizes the LFA concept of Penev and Atick (1996). We discuss the relation between our results, which are about regression, and the different problem of pattern classification.

Estimating Dependency Structure as a Hidden Variable

This paper introduces a probability model, the mixture of trees that can account for sparse, dynamically changing dependence relationships. We present a family of efficient algorithms that use EM and the Minimum Spanning Tree algorithm to find the ML and MAP mixture of trees for a variety of priors, including the Dirichlet and the MDL priors. We also show that the single tree classifier acts like an implicit feature selector, thus making the classification performance insensitive to irrelevant attributes. Experimental results demonstrate the excellent performance of the new model both in density estimation and in classification.

A Unified Framework for Regularization Networks and Support Vector Machines

Regularization Networks and Support Vector Machines are techniques for solving certain problems of learning from examples -- in particular the regression problem of approximating a multivariate function from sparse data. We present both formulations in a unified framework, namely in the context of Vapnik's theory of statistical learning which provides a general foundation for the learning problem, combining functional analysis and statistics.

Selecting Relevant Genes with a Spectral Approach

Array technologies have made it possible to record simultaneously the expression pattern of thousands of genes. A fundamental problem in the analysis of gene expression data is the identification of highly relevant genes that either discriminate between phenotypic labels or are important with respect to the cellular process studied in the experiment: for example cell cycle or heat shock in yeast experiments, chemical or genetic perturbations of mammalian cell lines, and genes involved in class discovery for human tumors. In this paper we focus on the task of unsupervised gene selection. The problem of selecting a small subset of genes is particularly challenging as the datasets involved are typically characterized by a very small sample size ?? the order of few tens of tissue samples ??d by a very large feature space as the number of genes tend to be in the high thousands. We propose a model independent approach which scores candidate gene selections using spectral properties of the candidate affinity matrix. The algorithm is very straightforward to implement yet contains a number of remarkable properties which guarantee consistent sparse selections. To illustrate the value of our approach we applied our algorithm on five different datasets. The first consists of time course data from four well studied Hematopoietic cell lines (HL-60, Jurkat, NB4, and U937). The other four datasets include three well studied treatment outcomes (large cell lymphoma, childhood medulloblastomas, breast tumors) and one unpublished dataset (lymph status). We compared our approach both with other unsupervised methods (SOM,PCA,GS) and with supervised methods (SNR,RMB,RFE). The results clearly show that our approach considerably outperforms all the other unsupervised approaches in our study, is competitive with supervised methods and in some case even outperforms supervised approaches.

Bit-Width Analysis for General Applications

It has been widely known that a significant part of the bits are useless or even unused during the program execution. Bit-width analysis targets at finding the minimum bits needed for each variable in the program, which ensures the execution correctness and resources saving. In this paper, we proposed a static analysis method for bit-widths in general applications, which approximates conservatively at compile time and is independent of runtime conditions. While most related work focus on integer applications, our method is also tailored and applicable to floating point variables, which could be extended to transform floating point number into fixed point numbers together with precision analysis. We used more precise representations for data value ranges of both scalar and array variables. Element level analysis is carried out for arrays. We also suggested an alternative for the standard fixed-point iterations in bi-directional range analysis. These techniques are implemented on the Trimaran compiler structure and tested on a set of benchmarks to show the results.

SCAPA (Spacially Addressable Protein Array) – a Novel Protein Array for the Differential Profiling of Ligand-receptor Induced Signalling

While protein microarray technology has been successful in demonstrating its usefulness for large scale high-throughput proteome profiling, performance of antibody/antigen microarrays has been only moderately productive. Immobilization of either the capture antibodies or the protein samples on solid supports has severe drawbacks. Denaturation of the immobilized proteins as well as inconsistent orientation of antibodies/ligands on the arrays can lead to erroneous results. This has prompted a number of studies to address these challenges by immobilizing proteins on biocompatible surfaces, which has met with limited success. Our strategy relates to a multiplexed, sensitive and high-throughput method for the screening quantification of intracellular signalling proteins from a complex mixture of proteins. Each signalling protein to be monitored has its capture moiety linked to a specific oligo ‘tag’. The array involves the oligonucleotide hybridization-directed localization and identification of different signalling proteins simultaneously, in a rapid and easy manner. Antibodies have been used as the capture moieties for specific identification of each signaling protein. The method involves covalently partnering each antibody/protein molecule with a unique DNA or DNA derivatives oligonucleotide tag that directs the antibody to a unique site on the microarray due to specific hybridization with a complementary tag-probe on the array. Particular surface modifications and optimal conditions allowed high signal to noise ratio which is essential to the success of this approach.

Tubular Organization of SnO₂ Nanocrystallites for Improved Lithium Ion Battery Anode Performance

Porous tin oxide nanotubes were obtained by vacuum infiltration of tin oxide nanoparticles into porous aluminum oxide membranes, followed by calcination. The porous tin oxide nanotube arrays so prepared were characterized by FE-SEM, TEM, HRTEM, and XRD. The nanotubes are open-ended, highly ordered with uniform cross-sections, diameters and wall thickness. The tin oxide nanotubes were evaluated as a substitute anode material for the lithium ion batteries. The tin oxide nanotube anode could be charged and discharged repeatedly, retaining a specific capacity of 525 mAh/g after 80 cycles. This capacity is significantly higher than the theoretical capacity of commercial graphite anode (372 mAh/g) and the cyclability is outstanding for a tin based electrode. The cyclability and capacities of the tin oxide nanotubes were also higher than their building blocks of solid tin oxide nanoparticles. A few factors accounting for the good cycling performance and high capacity of tin oxide nanotubes are suggested.

Diseño y realización de servicios de presentación en entornos gráficos.

En cada unidad didáctica precede al tít.: Técnico Superior en Desarrollo de Aplicaciones Informáticas y consta en marbetes: Formación Profesional a Distancia y Ciclo Formativo de Grado Superior

Identificación de escenarios a futuro para lograr competitividad en el sector de telefonía móvil en Colombia al año 2020

Este trabajo pretende explorar el desarrollo del sector de la telefonía móvil desde sus inicios hasta la actualidad en Colombia, con el fin de generar escenarios de futuro. Las herramientas prospectivas MicMac (Análisis Estructural Prospectivo), Smic (Sistema de Matrices de Impactos Cruzados) y la opinión de expertos líderes del sector, son la base principal para el desarrollo del trabajo. Las entidades gubernamentales, la CTR (Comisión de Regulación de Telecomunicaciones), y los líderes de los operadores del sector de telefonía móvil, entre otros, se han concientizado que la innovación es la base del éxito en este tipo de organizaciones y por eso se ha trabajado en mejorar su regulación, logrando de esta manera que el desarrollo de los productos y servicios que se ofrecen sean cada vez mejores y perjudique en menor medida al medio ambiente y a los usuarios. Este subsector de las telecomunicaciones, es el más dinámico y con mayor potencial. Sin embargo, este también es afectado por las condiciones económicas del mercado, la inestabilidad política, las importaciones y exportaciones derivadas de los tratados comerciales, entre otros temas. El escenario apuesta facilitaría la prestación de productos con tecnología de punta y servicios con la mejor cobertura y acceso posible a precios bajos.

Lecture 7 - Going Loopy

Looping while do while for Arrays indexes For each loop

Programming Principles: Collections and Iterators

In this session we look at how we can use collection objects like ArrayList as a more advanced type of array. We also introduce the idea of generics (forcing a collection to hold a particular type) and see how Java handles the autoboxing and unboxing of primitives. Finally we look at Iterators, a common design pattern for dealing with iteration over a collection.

Aspectos, consideraciones y orientaciones sobre la enseñanza del español de los negocios en Corea.

Resumen basado en el de la publicación. Memoria de master (Universidad Pablo de Olavide, 2010)

Las Juntas Neogranadinas y el constitucionalismo criollo pregaditano (Debates)

Este artículo plantea cómo las juntas de gobierno establecidas en el Nuevo Reino de Granada (Colombia), desengañadas por la posición de las Cortes de Cádiz frente a la representación y las aspiraciones criollas, y al mismo tiempo recelosas de las pretensiones hegemónicas de la capital virreinal, optaron por desarrollar un temprano y disperso proceso constituyente que dio origen a varias constituciones que antecedieron en el tiempo, y en más de una ocasión superaron en su concepción de la ciudadanía, el Estado y la sociedad a la Constitución de Cádiz en marzo de 1812.

Identification of Summer Origins of Songbirds Migrating through Southern Canada in Autumn

Stable hydrogen isotopes (δD) in flight feathers were measured to investigate the summer origins of five species of boreal-breeding warblers captured during fall migration at Canadian Migration Monitoring Network (CMMN) stations spread across southern Canada. Mean δD varied among stations and species within stations, but there was broad overlap in δD values. Although isotope ratios indicate that migrants at each station come from a wide range of latitudes, they are unable to provide much longitudinal discrimination. Band recoveries are sparse, but indicate that in general western Canadian warblers move southeast in fall, eastern birds move southwest, and there is a transition zone in the Great Lakes region. Combining knowledge of migratory direction with isotope results increases discrimination of breeding areas. Isotope results support fall migratory movements by Blackpoll Warbler (Dendroica striata) and Northern Waterthrush (Seiurus novaboracensis) that are more easterly than for other species, and in all study species, birds from more northern regions passed through southern Canada later in the season. Migration monitoring stations capture birds from broad areas of latitude, and migrants passing through each province appear to come from largely different portions of the Canadian breeding range, so a few stations placed in each province should suffice collectively to sample birds from most of the boreal forest. Migration monitoring in southern Canada, therefore, has the potential to monitor status of boreal forest birds in Canada that are unsampled by other monitoring programs.