Deep learning review and its applications

Deep neural networks have recently gained popularity for improv- ing state-of-the-art machine learning algorithms in diverse areas such as speech recognition, computer vision and bioinformatics. Convolutional networks especially have shown prowess in visual recognition tasks such as object recognition and detection in which this work is focused on. Mod- ern award-winning architectures have systematically surpassed previous attempts at tackling computer vision problems and keep winning most current competitions. After a brief study of deep learning architectures and readily available frameworks and libraries, the LeNet handwriting digit recognition network study case is developed, and lastly a deep learn- ing network for playing simple videogames is reviewed.

Uma análise dos esquemas de dígitos verificadores usados no Brasil

Neste trabalho discutimos vários sistemas de dígitos verificadores utilizados no Brasil, muitos deles semelhantes a esquemas usados mundialmente, e fazemos uma análise da sua capacidade de detectar os diversos tipos de erros que são comuns na entrada de dados em sistemas computacionais. A análise nos mostra que os esquemas escolhidos constituem decisões subotimizadas e quase nunca obtêm a melhor taxa de detecção de erros possível. Os sistemas de dígitos verificadores são baseados em três teorias da álgebra: aritmética modular, teoria de grupos e quasigrupos. Para os sistemas baseados em aritmética modular, apresentamos várias melhorias que podem ser introduzidas. Desenvolvemos um novo esquema ótimo baseado em aritmética modular base 10 com três permutações para identificadores de tamanho maior do que sete. Descrevemos também o esquema Verhoeff, já antigo, mas pouquíssimo utilizado e que também é uma alternativa de melhoria para identificadores de tamanho até sete. Desenvolvemos ainda, esquemas ótimos para qualquer base modular prima que detectam todos os tipos de erros considerados. A dissertação faz uso ainda de elementos da estatística, no estudo das probabilidades de detecção de erros e de algoritmos, na obtenção de esquemas ótimos.

A síndrome de burnout e suas representações entre profissionais de saúde de um hospital geral

A tese tem como objetivo geral analisar a ocorrência da síndrome de burnout e suas representações sociais entre profissionais de saúde de um hospital público universitário do Rio de Janeiro. Caracteriza-se como um estudo quanti-qualitativo, descritivo, transversal, a partir de uma amostra intencional, totalizando 101 participantes. Os instrumentos de coleta de dados foram um questionário contendo a técnica de evocações livres, a escala de caracterização do burnout (ECB) e um roteiro de entrevista semiestruturada. A análise de dados se deu por meio do uso do software EVOC 2005 e da construção do quadro de quatro casas para as evocações livres; o software SPSS 20 e análises estatísticas descritivas e inferenciais para a ECB; e a análise de conteúdo temático categorial para as entrevistas semidiretivas. Os resultados apontam que os profissionais de saúde possuem uma representação social da síndrome de burnout alicerçada em duas dimensões, uma física e outra psicológica; observa-se a prevalência de conteúdos predominantemente negativos nessa representação, sobretudo, em relação ao contexto de trabalho no hospital. Também mostram a existência da representação social da síndrome de burnout estruturada a partir dos termos estresse e cansaço que fazem parte do provável núcleo central dessa representação. Observou-se que há ocorrência da síndrome de burnout entre enfermeiros e médicos do campo cirúrgico do hospital e que tal ocorrência apresenta relação com as variáveis psicossociais e sóciodemográficas. Releva-se a importância dessas variáveis, assinalando o seu papel regulador na ocorrência e desenvolvimento da síndrome. Conclui-se destacando as condições de trabalho para a construção das representações nas profissões de saúde e para a determinação da síndrome de burnout, sobretudo no campo cirúrgico. Os resultados poderão contribuir para a compreensão do campo teórico da síndrome de burnout, especialmente, no que se refere à sua determinação e ao seu desenvolvimento, lançando luz para o seu diagnóstico e sua prevenção

As representações sociais da judicialização da saúde para profissionais de saúde

A judicialização da saúde é uma expressão cada vez mais presente nos cenários de assistência à saúde no Brasil materializada, principalmente, pelos mandados judiciais para a realização de procedimentos diagnósticos e terapêuticos, consultas, internações e dispensação de insumos médico-cirúrgicos. Como objetivo geral o estudo visa analisar o processo de judicialização da saúde no Brasil, no âmbito do SUS, a partir dos profissionais de saúde envolvidos nesse processo. Foram definidos como objetivos específicos descrever a estrutura e os conteúdos das representações sociais da judicialização da saúde no âmbito do SUS para gestores, reguladores e profissionais de saúde envolvidos no processo de atendimento às demandas geradas pela prática da ação judicial; comparar as Representações Sociais sobre a judicialização da saúde destes grupos sociais; discutir os conteúdos dessas representações no processo de judicialização da saúde compreendendo o campo social no qual esta questão se insere, as relações de poder que as perpassam e as construções simbólicas que as especificam. Trata-se de um estudo qualitativo, desenvolvido à luz da Teoria das Representações Sociais, realizado com 152 profissionais, em um hospital universitário e na central de regulação de procedimentos e leitos na cidade do Rio de Janeiro. Aplicou-se a Técnica de Evocação Livre através do termo indutor judicialização da saúde, sendo essas analisadas com a técnica de quadro de quatro casas e 40 entrevistas semiestruturadas analisadas através da análise de conteúdo temático-categorial instrumentalizada pelo software NVivo. Na análise estrutural foram evocadas 761 palavras, com média das ordens médias de evocação de três, com frequência máxima de 17 e mínima de 10, já a análise processual resultou em seis categorias, com 2257 unidades de registro, distribuídas em 85 temas. Identificou-se que os profissionais de saúde apresentam um posicionamento negativo diante da realidade imposta pela judicialização, entretanto reconhecem esse recurso como necessário mediante a crise da saúde pública brasileira. Os profissionais de saúde trabalham sob a pressão do poder judiciário, com a ameaça de prisão levando a um cotidiano estressante de suas práticas profissionais decorrente de uma ação ineficaz do Estado na execução da política de saúde. Conclui-se que a representação social encontra-se em fase final de consolidação. Considerando a representação social como determinante de práticas, concluímos também que as representações que emergiram deste estudo podem contribuir para a mudança das práticas dos profissionais de saúde que operacionalizam esse sistema. Espera-se desta forma, que ocorra a promoção de melhorias no resultado final da assistência direta aos usuários que buscam as instituições públicas de saúde para serem atendidos em suas necessidades de saúde, caracterizando-se dessa forma um desafio maior que caminha no sentido de fazer avançar a democracia e a cidadania.

O comércio intra-industrial brasileiro de 1990 a 2013: mensuração, decomposição e os determinantes

Em uma relação de bilateral,o comérciointra-industrial (CII)acontece quando os dois países exportam e importam produtos pertencentes a uma mesma indústria. Conforme apontam diversas pesquisas relacionadas a esse assunto, o intercâmbio intra-indústriade produtos tem aumentado sua participação nas transações bilaterais. Não obstante, os resultados mensurados para o comércio entre o Brasil e o resto do mundo revelam a predominância do engajamento inter-industrial em detrimento do intra-industrial. Observando dados de comércio que abrangem todos os produtos catalogados com seis dígitos no Sistema Harmonizado, esta dissertação investiga o comércio intra-industrial brasileiro entre o período de 1990 a 2013 sob suas duas formas: o intercâmbio de bens diferenciados horizontalmente e verticalmente. Ademais, a pesquisa de mensuração do CII vertical se dedica a definição dos lados superior e inferior, a fim de verificar se o Brasil é exportador ou importador líquido de bens verticalmente diferenciados de qualidade relativamente superior. Dentre a amostra de 57 países, as relações bilaterais em que o índice de Grubel e Lloyd (GL) se mostrou mais expressivo foram Argentina, Estados Unidos, México, Alemanha, Suécia, Uruguai, França, Itália, Reino Unido e Colômbia. Com relação aos setores que mais contribuíram para o índice que mensura o shareintra-industrial do comércio, destacam-se os ramos produtores de materiais de transporte e máquinas e materiais elétricos, também merecendo menção o setor de produtos químicos.Utilizando técnicas econométricas usuais da abordagem de dados em painel, a investigação dos determinantes do CII nos leva à aceitação da hipótese que relaciona o CII à presença de economias de escala. Além disso, os parâmetros estimados revelam que a distância geográfica que separa o Brasil de seu parceiro comercial prejudica mais o comércio intra-industrial vis-à-vis o inter-industrial. O processo de integração desencadeado pelo MERCOSUL, por sua vez, mostrou efeito positivo sobre o comércio intra-industrialvertical.

National Centers for Coastal Ocean Science Coastal Ecosystem Assessment Program: a manual of methods

Environmental managers strive to preserve natural resources for future generations but have limited decision-making tools to define ecosystem health. Many programs offer relevant broad-scale, environmental policy information on regional ecosystem health. These programs provide evidence of environmental condition and change, but lack connections between local impacts and direct effects on living resources. To address this need, the National Oceanic and Atmospheric Administration/National Ocean Service (NOAA/NOS) Cooperative Oxford Laboratory (COL), in cooperation with federal, state, and academic partners, implemented an integrated biotic ecosystem assessment on a sub-watershed 14-digit Hydrologic Unit Code (HUD) scale in Chesapeake Bay. The goals of this effort were to 1) establish a suite of bioindicators that are sensitive to ecosystem change, 2) establish the effects of varying land-use patterns on water quality and the subsequent health of living resources, 3) communicate these findings to local decision-makers, and 4) evaluate the success of management decisions in these systems. To establish indicators, three sub-watersheds were chosen based on statistical analysis of land-use patterns to represent a gradient from developed to agricultural. The Magothy (developed), Corsica (agricultural), and Rhode (reference) Rivers were identified. A random stratified design was developed based on depth (2m contour) and river mile. Sampling approaches were coordinated within this structure to allow for robust system comparisons. The sampling approach was hierarchal, with metrics chosen to represent a range from community to cellular level responses across multiple organisms. This approach allowed for the identification of sub-lethal stressors, and assessment of their impact on the organism and subsequently the population. Fish, crabs, clams, oysters, benthic organisms, and bacteria were targeted, as each occupies a separate ecological niche and may respond dissimilarly to environmental stressors. Particular attention was focused on the use of pathobiology as a tool for assessing environmental condition. By integrating the biotic component with water quality, sediment indices, and land- use information, this holistic evaluation of ecosystem health will provide management entities with information needed to inform local decision-making processes and establish benchmarks for future restoration efforts.

Kernelized log linear models for continuous speech recognition

Large margin criteria and discriminative models are two effective improvements for HMM-based speech recognition. This paper proposed a large margin trained log linear model with kernels for CSR. To avoid explicitly computing in the high dimensional feature space and to achieve the nonlinear decision boundaries, a kernel based training and decoding framework is proposed in this work. To make the system robust to noise a kernel adaptation scheme is also presented. Previous work in this area is extended in two directions. First, most kernels for CSR focus on measuring the similarity between two observation sequences. The proposed joint kernels defined a similarity between two observation-label sequence pairs on the sentence level. Second, this paper addresses how to efficiently employ kernels in large margin training and decoding with lattices. To the best of our knowledge, this is the first attempt at using large margin kernel-based log linear models for CSR. The model is evaluated on a noise corrupted continuous digit task: AURORA 2.0. © 2013 IEEE.

Continuous speech research based on HyperSausage Neuron

In this paper, we presents HyperSausage Neuron based on the High-Dimension Space(HDS), and proposes a new algorithm for speaker independent continuous digit speech recognition. At last, compared to HMM-based method, the recognition rate of HyperSausage Neuron method is higher than that of in HMM-based method.

Continuous speech research based on HyperSausage Neuron

In this paper, we presents HyperSausage Neuron based on the High-Dimension Space(HDS), and proposes a new algorithm for speaker independent continuous digit speech recognition. At last, compared to HMM-based method, the recognition rate of HyperSausage Neuron method is higher than that of in HMM-based method.

混合模式识别系统研究

讨论基于多种分类方法的模块组合实现的混合模式识别系统，它不同于利用多分类器输出结果表决的集成系统．提出两个系统：一个面向印刷体汉字文本识别，另一个面向自由手写体数字识别．利用多种特征和多种分类方法的组合、部分识别信息控制混淆字判别策略以及提出的动态模板库匹配后处理方法，使系统的性能与传统单一分类器系统比较，获得明显改善．实验表明：多方法多策略混合是解决复杂和增强系统鲁棒性的一条途径

提高勘探成熟区地震资料精度的方法研究

With the Oil field exploration and exploitation, the problem of supervention and enhaning combination gas recovery was faced.then proposing new and higher demands to precision of seismic data. On the basis of studying exploration status,resource potential,and quality of 3D seismic data to internal representative mature Oil field, taking shengli field ken71 zone as study object, this paper takes advantage of high-density 3D seismic technique to solving the complex geologic problem in exploration and development of mature region, deep into researching the acquisition, processing of high-density 3D seismic data. This disseration study the function of routine 3D seismic, high-density 3D seismic, 3D VSP seismic,and multi-wave multi-component seismic to solving the geologic problem in exploration and development of mature region,particular introduce the advantage and shortage of high-density 3D seismic exploration, put forward the integrated study method of giving priority to high-density 3D seismic and combining other seismic data in enhancing exploration accuracy of mature region. On the basis of detailedly studying acquisition method of high-density 3D seismic and 3D VSP seismic,aming at developing physical simulation and numeical simulation to designing and optimizing observation system. Optimizing “four combination” whole acquisition method of acquisition of well with ground seimic and “three synchron”technique, realizing acquisition of combining P-wave with S-wave, acquisition of combining digit geophone with simulation geophone, acquisition of 3D VSP seismic with ground seimic, acquisition of combining interborehole seismic,implementing synchron acceptance of aboveground equipment and downhole instrument, common use and synchron acceptance of 3D VSP and ground shots, synchron acquisition of high-density P-wave and high-density multi-wave, achieve high quality magnanimity seismic data. On the basis of detailedly analysising the simulation geophone data of high-density acquisition ,adopting pertinency processing technique to protecting amplitude,studying the justice matching of S/N and resolution to improving resolution of seismic profile ,using poststack series connection migration,prestack time migration and prestack depth migration to putting up high precision imaging,gained reliable high resolution data.At the same time carrying along high accuracy exploration to high-density digit geophone data, obtaining good improve in its resolution, fidelity, break point clear degree, interbed information, formation characteristics and so on.Comparing processing results ,we may see simulation geophone high-density acquisition and high precision imaging can enhancing resolution, high-density seismic basing on digit geophone can better solve subsurface geology problem. At the same time, fine processing converted wave of synchron acquisition and 3D VSP seismic data,acquiring good result. On the basis of high-density seismic data acquisition and high-density seismic data processing, carry through high precision structure interpretation and inversion, and preliminary interpretation analysis to 3D VSP seismic data and multi-wave multi-component seismic data. High precision interpretation indicates after high resolution processing ,structural diagram obtaining from high-density seismic data better accord with true geoligy situation.

中老年男性性激素水平与认知功能的关系

Abstract Objective: Study the relationship between sex hormones and cognitive function in aged male. Methods: The serum sex hormones including Free Testosterone(FT), Total Testosterone(TT), Estradiol(E2), Prolactin(PRL) and Luteotropic Hormone(LH) of 74 elderly men(47-75 years old, mean 58.73) were measured by radioimmunoassay. They were all applied the cognitive tests of episodic memory and verbal fluency. For the episodic memory, four experiments examined picture, digit and words memory separately. Four verbal fluency tests were performed in the middle of each episodic memory. The 74 men were divided into two groups according to the levels of their sex hormones respectively, with each group had 37 subjects. Data was analyzed by the software of SPSS 11.0. Results: Mean age and the mean years of formal education had no significant difference between the groups; Men in the group with higher levels of FT had significantly higher scores in the tests of picture memory and words memory；Men in the group with higher levels of TT had significantly higher scores in the tests of verbal fluency4, picture memory and words memory；Men in the group with higher levels of E2 had significantly higher scores in the tests of picture memory and verbal fluency3,4；There were no correlations between PRL or LH and the cognitive tests. Conclusions: 1 Sex hormones (especially FT, TT, E2) did have effect on the cognitive function in aged male, higher levels of hormones related with higher scores of some cognitive tests. 2 Among the cognitive functions, picture memory and words memory performance seemed being effected by the sex hormones most.

8～12岁男孩的性激素与智力、人格的关系

Recently there emerged more and more researhes concerning about sex hormone-related cognitive ability and behaviours. Few were carried out in pre- to early adolescent children., and the objective of the current study is to investigate whether there are covariance between sex hormones, intelligence and personality in 232 pre- to early adolescent boys, including 62 gifted boys. Indexes of sex hormone levels were salivary testosterone and estradiol concentrations, the 2D:4D digit ratio ( a reliable pointer of prenatal sex hormone concentraions). The Cattell Culture Fair Intelligence Test and a Chinese version of Children’s Personality Questionaire was applied in the current study. The main findings are: 1) salivary sex hormone concentrations significantly positively correlated with intelligence performance in 10-year-old boys; 2) salivary testosterone negatively related to intelligence performance in 12-year-old boys; 3) gifted boys bears lower testosterone concentrations in both prenatal period and pre- to early adolescence; 4) for personality, higher salivary estradiol was related to extraversion and digit ratios correlated with several personality factors in 8-year-old boys. In conclusion, results in the current study suggested that for male early maturers, intelligence may be negatively influenced by early coming androgen surge. In contrast, male late matures may benefit from their lately and moderately increasing hormone. Besides, the results suggested that the relationship between 2D:4D digit ratio and personality may also be paid attention to.

International transport costs and the margins of Intra-Latin American maritime trade.

This paper focuses on the analysis of the relationship between maritime trade and transport cost in Latin America. The analysis is based on disaggregated (SITC 5 digit level) trade data for intra Latin maritime trade routes over the period 1999-2004. The research contributes to the literature by disentangling the effects of transport costs on the range of traded goods (extensive margin) and the traded volumes of goods (intensive margin) of international trade in order to test some of the predictions of the trade theories that introduce firm heterogeneity in productivity, as well as fixed costs of exporting. Recent investigations show that spatial frictions (distance) reduce trade mainly by trimming the number of shipments and that most firms ship only to geographically proximate customers, instead of shipping to many destinations in quantities that decrease in distance. Our analyses confirm these findings and show that the opposite pattern is observed for ad-valorem freight rates that reduce aggregate trade values mainly by reducing the volume of imported goods (intensive margin).

Simultaneous Localization and Recognition of Dynamic Hand Gestures

A framework for the simultaneous localization and recognition of dynamic hand gestures is proposed. At the core of this framework is a dynamic space-time warping (DSTW) algorithm, that aligns a pair of query and model gestures in both space and time. For every frame of the query sequence, feature detectors generate multiple hand region candidates. Dynamic programming is then used to compute both a global matching cost, which is used to recognize the query gesture, and a warping path, which aligns the query and model sequences in time, and also finds the best hand candidate region in every query frame. The proposed framework includes translation invariant recognition of gestures, a desirable property for many HCI systems. The performance of the approach is evaluated on a dataset of hand signed digits gestured by people wearing short sleeve shirts, in front of a background containing other non-hand skin-colored objects. The algorithm simultaneously localizes the gesturing hand and recognizes the hand-signed digit. Although DSTW is illustrated in a gesture recognition setting, the proposed algorithm is a general method for matching time series, that allows for multiple candidate feature vectors to be extracted at each time step.