Advances in scaling deep learning algorithms

Les algorithmes d'apprentissage profond forment un nouvel ensemble de méthodes puissantes pour l'apprentissage automatique. L'idée est de combiner des couches de facteurs latents en hierarchies. Cela requiert souvent un coût computationel plus elevé et augmente aussi le nombre de paramètres du modèle. Ainsi, l'utilisation de ces méthodes sur des problèmes à plus grande échelle demande de réduire leur coût et aussi d'améliorer leur régularisation et leur optimization. Cette thèse adresse cette question sur ces trois perspectives. Nous étudions tout d'abord le problème de réduire le coût de certains algorithmes profonds. Nous proposons deux méthodes pour entrainer des machines de Boltzmann restreintes et des auto-encodeurs débruitants sur des distributions sparses à haute dimension. Ceci est important pour l'application de ces algorithmes pour le traitement de langues naturelles. Ces deux méthodes (Dauphin et al., 2011; Dauphin and Bengio, 2013) utilisent l'échantillonage par importance pour échantilloner l'objectif de ces modèles. Nous observons que cela réduit significativement le temps d'entrainement. L'accéleration atteint 2 ordres de magnitude sur plusieurs bancs d'essai. Deuxièmement, nous introduisont un puissant régularisateur pour les méthodes profondes. Les résultats expérimentaux démontrent qu'un bon régularisateur est crucial pour obtenir de bonnes performances avec des gros réseaux (Hinton et al., 2012). Dans Rifai et al. (2011), nous proposons un nouveau régularisateur qui combine l'apprentissage non-supervisé et la propagation de tangente (Simard et al., 1992). Cette méthode exploite des principes géometriques et permit au moment de la publication d'atteindre des résultats à l'état de l'art. Finalement, nous considérons le problème d'optimiser des surfaces non-convexes à haute dimensionalité comme celle des réseaux de neurones. Tradionellement, l'abondance de minimum locaux était considéré comme la principale difficulté dans ces problèmes. Dans Dauphin et al. (2014a) nous argumentons à partir de résultats en statistique physique, de la théorie des matrices aléatoires, de la théorie des réseaux de neurones et à partir de résultats expérimentaux qu'une difficulté plus profonde provient de la prolifération de points-selle. Dans ce papier nous proposons aussi une nouvelle méthode pour l'optimisation non-convexe.

Biochemical & Pharmacological evaluation of liver oils of selected deep sea sharks and chimaeras of the Indian EEZ

With a seacoast of 8,1 18 km, an exclusive economic zone (EEZ) of 2 million square km, and with an area of about 30,000 square km under aquaculture, lndia produces close to six million tonnes of fish, over 4 per cent of the world fish production. While the marine waters upto 50m depth have been fully exploited, those beyond, remain unexplored. There is an ever increasing demand for fishery resources as food. The coastal fishery resources of the country are dwindling at a rapid pace and it becomes highly imperative that we search for alternate fishery resources for food. The option we have is to hunt for marine fishery resources. Studies pertaining to proximate composition, amino acid and fatty acid composition are essential to understand the nutraceutical values of these deep sea fishery resources. The present study was aimed to carry out proximate composition of deep sea fishery resources obtained during cruises onboard the FORV Sarise Sampada, to identify fishery resources which have appreciable lipid content and thereby analyse the bioactive potentials of marine lipids, to study the amino acid profile of these fishery resources, to understand the contents of SPA, MUFA and PUFA and to calculate the n3/n6 fatty acid contents. Though the presence of nutraceuticals was identified in the marine fishery resources their use as potential food resources deserve further investigation. So the study were carried out to calculate the hepatosomatic indices of sharks & chimaeras and conduct biochemical characterisation of liver oils of Apristurus indicus, Cenlrophorus scalprams, Centroselachus crepidater, Neoharriotta raleighana, and Harriotta pinnata obtained during cruises onboard the FORV Sugar Sampada.Therapeutic use of shark liver oil is evident from its use for centuries as a remedy to heal wounds and fight flu (Neil er al. 2006). Japanese seamen called it 'samedava' or "cure all". Shark liver oil is being promoted worldwide as a dietary supplement to boost the immune system, fight infections, to treat cancer and to lessen the side effects of conventional cancer treatment. These days more emphasis is laid on the nutritive benefits of shark liver oils especially on the omega 3 polyunsaturated fatty acids ( PUFAs) (Anandan er al. 2007) and alkylglycerols (AKGs) (Pugliese er al. I998) contained in them due to the high rise of inflammatory disorders such as arthritis, asthma and neurodegenerative diseases like Alzheimer’s, Parkinson’s and Schizophrenia. So the present study also evaluate the pharmacological properties with respect to analgesic, anti-inflammatory, anti pyretic and anti-ulcer effects of four different liver oils of sharks belonging to the Indian EEZ and to identify the components of oil responsible for these activities.The analgesic and anti-inflammatory activities of liver oils from Neoharriotra raleighana (NR), Centrosymnus crepidater (CC), Apristurus indicus (AI), and Centrophorus sculpratus (CS) sharks caught from the Arabian Sea and the Indian Ocean were compared. The main objectives also include determination of the cholesterol lowering effects of liver oils of Neoharriotra raleighana (NR) and Centrophorus sculpratus (CS) on the high fat diet induced dyslipidemia and to compare the impact of four isolipidemic diets, on levels of serum diagnostic marker enzymes, on lipid profile of blood and liver and antioxidant status of heart in male Albino rats. And also to study the efficacy of Centrophorus sculpratus (CS) liver oil against Complete Freund’s Adjuvant-induced arthritis and to compare the anti-inflammatory activity of this oil with a traditionally used anti-inflammatory substance gingerol (oleoresin extracted from ginger.). The results of the present study indicated that both (Centrophorus sculpratus liver oils as well as gingerol extracts proved to be effective natural remedies against CFA-induced arthritis in Albino rats.

Systematics of properties of the electron gas in deep-etched quantum wires

An efficient method is developed for an iterative solution of the Poisson and Schro¿dinger equations, which allows systematic studies of the properties of the electron gas in linear deep-etched quantum wires. A much simpler two-dimensional (2D) approximation is developed that accurately reproduces the results of the 3D calculations. A 2D Thomas-Fermi approximation is then derived, and shown to give a good account of average properties. Further, we prove that an analytic form due to Shikin et al. is a good approximation to the electron density given by the self-consistent methods.

Influencia de las creencias religiosas cristianas en la decisión de realizar suicidio asistido al final de la vida

El suicidio asistido como una posible opción al final de la vida, es una idea que hasta ahora está siendo considerada, ya que existen argumentaciones a favor y en contra que han generado controvertidos debates a su alrededor. Algunos de los argumentos en contra están basados en los principios de las instituciones religiosas de orden cristiano, las cuales defienden el valor sagrado de la vida de las personas y la aceptación del sufrimiento como un acto de amor profundo y sumisión a los mandatos de Dios, el creador. Mientras del lado contrario, se encuentran quienes defienden el procedimiento, impulsando la autonomía y la autodeterminación que cada persona tiene sobre su vida. La revisión de la literatura realizada no sólo permite ampliar los argumentos de estas dos posiciones, sino que también permite conocer la historia del suicidio asistido, la posición que este procedimiento tiene en diferentes países del mundo, incluyendo a Colombia, y finalmente se presentan las contribuciones de la psicología entorno al procedimiento en discusión.

Sharks and other monsters of the deep.

Los océanos están considerados como la gran reserva de vida del planeta. Aquí se relata la vida de los tiburones y algunos otros depredadores marinos que cazan en todos los mares, con el objetivo de interesar al niño en saber más sobre el mundo en que vive. En las últimas páginas se hace un llamamiento para salvar al mar de la contaminación producida por el ser humano y salvar a muchas criaturas marinas que están amenazadas de extinción.

Beyond Student Voice : Patterns of Partnership and the Demands of Deep Democracy. 'Más allá de la voz del alumnado : patrones de colaboración y las exigencias de la democracia profunda'.

Monográfico con el título: 'Las posibilidades de la voz del alumnado para el cambio y la mejora educativa'. Resumen basado en el de la publicación

L'Emmagatzematge de cereals en sitges d'època ibèrica al nord-est de Catalunya

Una sitja és una cavitat subterrània destinada a emmagatzemar la collita, especialment de cereals. Amb el manteniment d'unes condicions ideal de temperatura i humitat els cereals s'hi poden conservar durant un llarg període de temps, que segons Varró podria arribar als 50 anys. Aquestes excepcionals possibilitats han possibilitat que l'emmagatzematge en sitges fos un dels mètodes de conservació de cereals a llarg termini més utilitzat en les societats pre-industrials de tot el món. La sitja estàndard del nord-est de Catalunya és aquella que era excavada a l'argila, no portava revestiment i tenia la boca en forma de tub, de 0,77 m de diàmetre màxim per 0,42 de profunditat. El perfil era de tipus còncau, amb el diàmetre màxim situat en el terç central de la sitja, i un fons indistintament còncau o pla. La profunditat i el diàmetre màxim es situarien entre 1,75 i 2 m., amb un marge de diferència reduïdíssim entre ambdues mesures. La capacitat resultant d'aquestes dimensions se situaria entre 1 i 3 tones de cereals, que en termes estàndards de producció seria el resultat de la collita d'una extensió d'entre 1,5 i 4 hectàrees de terreny. ASBTRACT: A silo is an underground cavity designed to store the harvest, especially grain. With the maintenance of ideal conditions of temperature and moisture grains can be preserved for a long period of time, according to Varró it could reach 50 years. These exceptional opportunities have enabled the storage silos to be one of the methods of long-term conservation of grain used in most pre-industrial societies around the world. The standard silo from the North-East of Catalonia was excavated in clay,it had no siding and its mouth was tube-shaped, up to 0.77 m of maximum diameter to 0.42 deep. The profile was concave, with maximum diameter located in the central third of the silo, and a background either concave or flat. The depth and maximum diameter are located between 1.75 and 2 m, with a very little margin of difference between the two measures. The capacity resulting from these dimensions would be located between 1 and 3 tons of cereals, which in terms of production standards it would mean a harvest of between 1.5 and 4 hectares of ground.

Comentarios al proceso de integración europea desde una perspectiva migratoria latinoamericana. La Unión Europea frente a los inmigrantes no comunitarios

Este artículo analiza cómo el proceso de integración de la Unión Europea, por un lado, favorece a la libre circulación de mercaderías, capitales y de ciudadanos comunitarios, bajo la perspectiva de un sistema de integración profundo, y, por otro, paulatinamente ha dejado de lado a los denominados “otros no comunitarios”, a través de las denominadas “directivas de retorno” y la implementación de medidas tendientes hacia la criminalización de la migración, lo cual refleja una vulneración a los derechos fundamentales de la población inmigrante contrariando la tendencia moderna hacia una ciudadanía universal.

Simulated last glacial maximum D14Catm and the deep glacial ocean carbon reservoir

∆14Catm has been estimated as 420 ± 80‰ (IntCal09) during the Last Glacial Maximum (LGM) compared to preindustrial times (0‰), but mechanisms explaining this difference are not yet resolved. ∆14Catm is a function of both cosmogenic production in the high atmosphere and of carbon cycling and partitioning in the Earth system. 10Be-based reconstructions show a contribution of the cosmogenic production term of only 200 ± 200‰ in the LGM. The remaining 220‰ have thus to be explained by changes in the carbon cycle. Recently, Bouttes et al. (2010, 2011) proposed to explain most of the difference in pCO2atm and δ13C between glacial and interglacial times as a result of brine-induced ocean stratification in the Southern Ocean. This mechanism involves the formation of very saline water masses that contribute to high carbon storage in the deep ocean. During glacial times, the sinking of brines is enhanced and more carbon is stored in the deep ocean, lowering pCO2atm. Moreover, the sinking of brines induces increased stratification in the Southern Ocean, which keeps the deep ocean well isolated from the surface. Such an isolated ocean reservoir would be characterized by a low ∆14C signature. Evidence of such 14C-depleted deep waters during the LGM has recently been found in the Southern Ocean (Skinner et al. 2010). The degassing of this carbon with low ∆14C would then reduce ∆14Catm throughout the deglaciation. We have further developed the CLIMBER-2 model to include a cosmogenic production of 14C as well as an interactive atmospheric 14C reservoir. We investigate the role of both the sinking of brine and cosmogenic production, alongside iron fertilization mechanisms, to explain changes in ∆14Catm during the last deglaciation. In our simulations, not only is the sinking of brine mechanism consistent with past ∆14C data, but it also explains most of the differences in pCO2atm and ∆14Catm between the LGM and preindustrial times. Finally, this study represents the first time to our knowledge that a model experiment explains glacial-interglacial differences in pCO2atm, δ13C, and ∆14C together with a coherent LGM climate.

Sudden stratospheric warmings seen in MINOS deep underground muon data

The rate of high energy cosmic ray muons as measured underground is shown to be strongly correlated with upper-air temperatures during short-term atmospheric (10-day) events. The effects are seen by correlating data from the MINOS underground detector and temperatures from the European Centre for Medium Range Weather Forecasts during the winter periods from 2003-2007. This effect provides an independent technique for the measurement of meteorological conditions and presents a unique opportunity to measure both short and long-term changes in this important part of the atmosphere. Citation: Osprey, S., et al. (2009), Sudden stratospheric warmings seen in MINOS deep underground muon data, Geophys. Res. Lett., 36, L05809, doi: 10.1029/2008GL036359.

Comparison between blue and green stimulated luminescence of Al(2)O(3):C

This paper presents a systematic comparison of OSL signals from Al(2)O(3):C when stimulated with blue and green light. Al(2)O(3):C detectors were irradiated with various doses and submitted to various bleaching regimes using yellow, green and blue light. Most of the investigations were carried out using Luxel (TM)-type detectors used in the commercial Luxet (TM) and InLight (TM) dosimetry systems (Landauer Inc.). Al(2)O(3):C single crystals and Al(2)O(3):C powder were also used to complement the investigations. The results show that, although blue stimulation provides faster readout times (OSL curves that decayed faster) and higher initial OSL intensity than green stimulation, blue stimulation introduced complicating factors. These include incomplete bleaching of the dosimetric trap when the Al(2)O(3):C detectors are bleached with yellow or green light and the OSL is recorded with blue light stimulation, and an increased residual level due to stimulation of charge carriers from deep traps. The results warrant caution when using blue stimulation to measure the OSL signal from Al(2)O(3):C detectors, particularly if the doses involved are low and the detectors have been previously exposed to high doses. (C) 2010 Elsevier Ltd. All rights reserved.

Avaliação do modelo de predição clínica de Wells et al. no diagnóstico da trombose venosa profunda dos membros inferiores

CONTEXTO: A aplicação de uma estratégia baseada em um modelo clínico associado ao mapeamento dúplex (MD) pode permitir um diagnóstico da trombose venosa profunda (TVP) mais seguro, eficaz e custo-efetivo. OBJETIVO: Testar o modelo clínico de Wells et al. associado ao MD e verificar a ocorrência de TVP nos pacientes categorizados quanto à probabilidade de apresentar a doença, e determinar se, a partir dos resultados obtidos, seria possível reduzir o número de exames seriados com o MD. MÉTODOS: Os pacientes com suspeita clínica de TVP foram categorizados quanto à apresentação de TVP em baixa, moderada e alta probabilidade (BP, MP, AP) e, em seguida, submetidos ao MD. Pacientes com MD negativo repetiram o exame em 24-48 horas e em 7 dias. Pacientes com exame positivo para TVP foram tratados. Todos os pacientes sem TVP foram convocados para reavaliação clínica em 3 meses. RESULTADOS: A ocorrência de TVP entre os 489 pacientes avaliados foi de 39,1% (191), sendo 35,6% identificados no exame inicial e 3,5% no exame seriado. Os índices de pacientes que apresentaram TVP foram de 6,1% no grupo de BP, 26,9% no grupo de MP e 79,5% no grupo de AP. No exame seriado, o percentual de TVP foi de 2,4, 7,8 e 15,1% nos grupos BP, MP e AP, respectivamente. Dos pacientes com MD negativo, 62,4% compareceram após 3 meses, e piora dos sintomas foi apresentada por apenas um paciente. Neste, o MD mostrou TVP de veia poplítea. CONCLUSÃO: Os resultados obtidos sugerem que, para os pacientes com BP para TVP e MD negativo, seria possível prescindir do exame seriado, devido à baixa ocorrência de TVP neste grupo, tornando, assim, a abordagem diagnóstica mais simples.

Morphological analysis of pits formed on Al 2024-T3 in chloride aqueous solution

It has been used a new image analysis method, based on segmentation by shape parameters, for pits morphology examination from Al 2024 aluminum-copper alloy in chloride aqueous solution. Corrosion behavior of this alloys in naturally aerated 3.5% NaCl solution has been investigated through open circuit potential measurements. Afterwards, pits have been characterized by image analysis taking density and size measurements right from corroded surfaces. Morphological investigation has been conducted for profiles, cut orthogonally from mean surface planes, and observed through light microscopy. Image analysis data could demonstrate that pits are wider than deep, evoluting for conical, quasi-conical or irregular shapes. Most pits have presented a quasi-conical morphology, but the wider ones have evoluted to an irregular shape influenced by sub-surface microstructure. Image analysis based on shape segmentation could enhance the differences on morphological behavior. (C) 2004 Elsevier B.V. All rights reserved.

Variaciones del índice de cono en suelo movilizado posterior al tráfico de un tractor equipado con dos tipos de neumáticos y a cuatro velocidades de desplazamiento

The tractor is one of the machines that more traffics over the soil during the processes involving agricultural production. The interaction tractor/soil is made by the tires which, in most of the cases, are pneumatic. The tire type and the tractor travel speed, interfere directly on the pressure over the soil. One of the techniques employed to evaluate the alterations that tractor traffic causes in the soil is to measure its Cone Index. The aim of this research was to evaluate the same Cone Index alterations caused by an agricultural tractor equipped with both radial tires and bias ply tires, trafficking mobilized soil in four different travel speeds. The experiment was performed in a LATOSSOLO VERMELHO, located 22°51' S, 48°25'W and 770 m of altitude, in Botucatu-SP, Brazil. The soil mobilization was performed with a chisel plow and a disc arrow. The traction was accomplished with a John Deere tractor, model 6600, with 88 kW of power and 6,723 kg. Equipment requiring a force of 25kN was traced by the tractor draw bar. The experimental design was in randomized blocks with 4 × 2 factorial arrangements, with two distinct treatments corresponding to the types of tires (bias and radial) and the four travel speeds, with six replications. There were selected the following speeds: 3.5, 3.9, 5.1 and 5.9 km h-1. To determine the soil resistance, there was utilized MSSU - Mobile Soil Sampling Unit, with which the Cone Index was obtained in layers from 0-100, 100-200, 200-300, 300-400, 400-500 and over 500 mm deep. The Cone Index where evaluated in areas with non contact between tire and soil (ICn) and in the tire footprint track (ICp). There were calculated the Cone Index increments caused by the tractor tire (AIC) and the results showed that as the tractor travel speed increased, there were observed decrements in the medium values of cone index. The radial tire provided smaller values of the Cone Index in the superficial layer of the soil (0 to 100 mm) in relation to the bias ply tire, when the speed was approximately 6 km h-1. The increment in the Cone Index, promoted by the tractor, was more intense in the first 200 mm depth, but it also reached the layer from 200 to 300 mm.

Diagnóstico da trombose venosa profunda dos membros inferiores, utilizando o modelo clínico de Wells et al. (2003), Dímero-D, Mapeamento Dúplex e avaliação da Proteina C Reativa

Pós-graduação em Bases Gerais da Cirurgia - FMB