Search

Acquiring and speaking a second language increases demand on the processes of language control for bilingual as compared to monolingual speakers. Language control for bilingual speakers involves the ability to keep the two languages separated to avoid interference and to select one language or the other in a given conversational context. This ability is what we refer with the term “bilingual language control” (BLC). It is now well established that the architecture of this complex system of language control encompasses brain networks involving cortical and subcortical structures, each responsible for different cognitive processes such as goal maintenance, conflict monitoring, interference suppression, and selective response inhibition. Furthermore, advances have been made in determining the overlap between the BLC and the nonlinguistic executive control networks, under the hypothesis that the BLC processes are just an instantiation of a more domain‐general control system. Here, we review the current knowledge about the neural basis of these control systems. Results from brain imaging studies of healthy adults and on the performance of bilingual individuals with brain damage are discussed. The research was supported by grants from the Spanish government (PSI2014-52181-P, and PSI2017-87784-R), the Catalan government (SGR 2009-1521, 2017SGR268), the La Marató de TV3 Foundation (373/C/2014), and the European Research Council under the European Community’s Seventh Framework (FP7/ 2007–2013 Cooperation Grant Agreement 613465-AThEME). Marco Calabria was supported by the postdoctoral Ramón y Cajal fellowship (RYC2013-14013).

Al present projecte s’han desenvolupat una sèrie de mesures i aplicacions pròpies de l’enginyeria elèctrica per a optimitzar energèticament una vivenda unifamiliar. S’ha seguit la filosofia descrita a la normativa ISO 50001, que estableix els requisits que ha de tenir un Sistema de Gestió Energètica. És important la creació d’una política energètica ben marcada cercant sempre l’ús més òptim dels recursos energètics que es tenen i encaminant la instal·lació cap a la transició energètica que tant es cerca avui en dia. Un dels trets més importants de la política energètica ha estat la intenció de cercar sempre una millora contínua de la instal·lació. Per això mateix, els recursos energètics distribuïts i les energies renovables han pres una gran importància pel que fa a la generació elèctrica de l’habitatge. L’energia solar fotovoltaica és la gran protagonista. Per a introduir els recursos energètics distribuïts es dissenya una microxarxa híbrida amb emmagatzematge d’energia. Les fonts generadores són la instal·lació fotovoltaica i la xarxa elèctrica de distribució. S’han tingut en compte els efectes mediambientals, d’estalvi energètic i d’estalvi econòmic d’igual manera, cercant sempre la solució més òptima. Per concloure, s’ha aconseguit reduir un 50 % la despesa econòmica referent a la facturació elèctrica anual.

Submitted by Divis??o de Documenta????o/BC Biblioteca Central (ddbc@ufam.edu.br) on 2018-03-02T13:31:20Z No. of bitstreams: 2 Disserta????o_Albert F. J. Costa.pdf: 4215192 bytes, checksum: 995a811676e714bffa60b5d73a387cfa (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Approved for entry into archive by Divis??o de Documenta????o/BC Biblioteca Central (ddbc@ufam.edu.br) on 2018-03-02T13:31:36Z (GMT) No. of bitstreams: 2 Disserta????o_Albert F. J. Costa.pdf: 4215192 bytes, checksum: 995a811676e714bffa60b5d73a387cfa (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Made available in DSpace on 2018-03-02T13:31:36Z (GMT). No. of bitstreams: 2 Disserta????o_Albert F. J. Costa.pdf: 4215192 bytes, checksum: 995a811676e714bffa60b5d73a387cfa (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-12-11 FAPEAM - Funda????o de Amparo ?? Pesquisa do Estado do Amazonas Current machine learning techniques used for prediction tasks rely on the assumption that the environment where data is generated is static and supervised. However, most of the real-world problems present dynamic and semi-supervised environments, which invalidate this assumed assumption. In these environments, it is possible the occurrence of a phenomenon known in the literature as concept drift. The main characteristic of such a phenomenon is that the relationship between instances??? features and their true classes changes over time. The primary effect of the concept drift occurrence is the decrease on classifier performance, which leads the classifier to be obsolete for the task. There are methods in the literature that deal with concept drift implicitly or explicitly. The main drawback to blind (implicit) methods is the cost on retraining the classifier constantly, even with no concept drift occurrence, while the informed (explicit) methods generally work based on classifier performance decrease, or based on assumptions about the samples distribution. In order to try to overcome these drawbacks, in this work we propose to use density variation of the most significant instances as an explicit trigger for concept drift detection. Density variation measurement is based on Active Learning, and is calculated from virtual margins projected onto the input space according to the classifier confidence. Focusing on demonstrating the validity of the proposed method, called Concept Drift Detection Method Based on Active Learning (DMAA), were have carried out experiments divided into two series. In the first series, DMAA was investigated on six databases, which represent four synthetic and two real problems. The obtained results show that the proposed method achieved 95.45% of drift detection rate on synthetic databases, and 16.5% as mean error rate in both synthetic and real databases. In the second series of experiments, DMAA was compared to three baselines, including two supervised drift detectors and one Active Learning-based method. The obtained results show that DMAA reached statistically significant better recognition rates in the majority of databases. Besides, it reduced the amount of labeled instances needed to keep the system updated. As atuais t??cnicas de aprendizado de m??quina aplicadas na tarefa de predi????o s??o baseadas na premissa de que o ambiente em que os dados s??o gerados apresenta um comportamento estacion??rio e supervisionado. Por??m, os ambientes, na maioria dos problemas do mundo real, s??o considerados din??micos e semi-supervisionados, fatos que invalidam as premissas normalmente utilizadas. Nesses ambientes h?? a possibilidade da ocorr??ncia do fen??meno conhecido na literatura por mudan??a de conceito (do ingl??s concept drift), que caracteriza-se pela altera????o na rela????o entre as caracter??sticas das inst??ncias e a sua verdadeira classe com a passagem do tempo. Como efeito prim??rio da ocorr??ncia desse fen??meno tem-se a degrada????o significativa na taxa de desempenho do classificador, tornando-o obsoleto para a tarefa. Encontram-se na literatura m??todos que lidam implicitamente ou explicitamente com a mudan??a de conceito, sendo que os m??todos cegos (impl??citos) arcam com os custos de retreinar o classificador de forma constante, enquanto que os m??todos informados (expl??citos) atuam normalmente baseados no monitoramento da degrada????o do desempenho do classificador, ou na realiza????o de suposi????es sobre a distribui????o das inst??ncias. Para contornar essas dificuldades, esta disserta????o prop??e o uso da varia????o da densidade das inst??ncias mais significativas, calculada com base em Aprendizado Ativo, como sinalizador expl??cito da ocorr??ncia de mudan??a de conceito. A densidade ?? mensurada a partir do conceito de margens virtuais projetadas no espa??o de entrada, sendo que as margens virtuais s??o obtidas com base na incerteza do classificador. Objetivando-se demonstrar a validade do m??todo proposto, denominado de M??todo de Detec????o de Mudan??a de Conceito Baseada em Aprendizado Ativo (DMAA), experimentos foram realizados em duas etapas. A primeira consistiu na aplica????o do DMAA em seis bases de dados, sendo quatro sint??ticas e duas reais. Os resultados obtidos demonstram que o m??todo proposto identificou em m??dia 95,45% das mudan??as existentes nas bases sint??ticas, e alcan??ou uma m??dia geral de erro de 16,5%. Na segunda etapa de experimentos, foi feita uma compara????o entre o DMAA e tr??s baselines, incluindo dois m??todos supervisionados e um m??todo baseado em aprendizado ativo. Os resultados indicam que o DMAA alcan??ou resultados estatisticamente superiores em mais da metade das bases de dados investigadas, al??m de reduzir significativamente a quantidade de inst??ncias rotuladas necess??rias para manter o sistema atualizado.

Published: October 3, 2017 All relevant data are available from the Github database at https:// github.com/sariac/Pupil-sizes_-Foreign-versus- Native.git. All relevant data are available from the Github database at https:// github.com/sariac/Pupil-sizes_-Foreign-versus- Native.git. Foreign languages are often learned in emotionally neutral academic environments which differ greatly from the familiar context where native languages are acquired. This difference in learning contexts has been argued to lead to reduced emotional resonance when confronted with a foreign language. In the current study, we investigated whether the reactivity of the sympathetic nervous system in response to emotionally-charged stimuli is reduced in a foreign language. To this end, pupil sizes were recorded while reading aloud emotional sentences in the native or foreign language. Additionally, subjective ratings of emotional impact were provided after reading each sentence, allowing us to further investigate foreign language effects on explicit emotional understanding. Pupillary responses showed a larger effect of emotion in the native than in the foreign language. However, such a difference was not present for explicit ratings of emotionality. These results reveal that the sympathetic nervous system reacts differently depending on the language context, which in turns suggests a deeper emotional processing when reading in a native compared to a foreign language. This research was partially supported by grants from the Spanish Government (PSI2014-52181-P, PSI2015-65689-P, PSI2011-23033 and SEV-2015-0490) (http://www.mineco.gob.es/portal/site/mineco/), from the Catalan Government (SGR-2014SGR1210), from the 7th Framework Programme (AThEME 613465) (http://www.atheme.eu/), and a 2016 BBVA Foundation Grant for Researchers and Cultural Creators (http://www.fbbva.es/) awarded to JAD.