Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 6
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Semantic and syntactic information for neural machine translation: Injecting features to the transformer

  • Armengol Estapé, Jordi
  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
Introducing factors such as linguistic features has long been proposed in machine translation to improve the quality of translations. More recently, factored machine translation has proven to still be useful in the case of sequence-to-sequence systems. In this work, we investigate whether this gains hold in the case of the state-of-the-art architecture in neural machine translation, the Transformer, instead of recurrent architectures. We propose a new model, the Factored Transformer, to introduce an arbitrary number of word features in the source sequence in an attentional system. Specifically, we suggest two variants depending on the level at which the features are injected. Moreover, we suggest two combination mechanisms for the word features and words themselves. We experiment both with classical linguistic features and semantic features extracted from a linked data database, and with two low-resource datasets. With the best-found configuration, we show improvements of 0.8 BLEU over the baseline Transformer in the IWSLT German-to-English task. Moreover, we experiment with the more challenging FLoRes English-to-Nepali benchmark, which includes both low-resource and very distant languages, and obtain an improvement of 1.2 BLEU. These improvements are achieved with linguistic and not with semantic information., This work is supported by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant Agreement No. 947657)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Multilingual machine translation: Deep analysis of language-specific encoder-decoders

  • Escolano Peinado, Carlos|||0000-0001-6657-673X
  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
  • Rodríguez Fonollosa, José Adrián|||0000-0001-9513-7939
State-of-the-art multilingual machine translation relies on a shared encoder-decoder. In this paper, we propose an alternative approach based on language-specific encoder-decoders, which can be easily extended to new languages by learning their corresponding modules. To establish a common interlingua representation, we simultaneously train N initial languages. Our experiments show that the proposed approach improves over the shared encoder-decoder for the initial languages and when adding new languages, without the need to retrain the remaining modules. All in all, our work closes the gap between shared and language-specific encoder-decoders, advancing toward modular multilingual machine translation systems that can be flexibly extended in lifelong learning settings., This work is supported by the European Research Council (ERC) under the European Union’sHorizon 2020 research and innovation programme (grant agreement No. 947657)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Converses al voltant de la intel·ligència artificial en clau catalana, Conversations around artificial intelligence from the catalan perspective

  • Ruiz Costa-Jussà, Marta|||0000-0002-5703-520X
  • Melero Nogues, Maite
La intel·ligència artificial mou milions d’euros i ocupa una part important de les agendes polítiques i estratègiques dels governs. En aquest article reflexionem sobre aquest concepte difús i ho fem a través de tres entrevistes a tres persones que ocupen llocs rellevants en aquest àmbit, i que a més tenen en comú que són nascudes i formades a Catalunya. Amb elles hem parlat distesament dels orígens d’aquesta revolució tecnològica, del seu impacte sobre l’organització social i el mercat laboral, de les seves implicacions ètiques, de les tendències futures que es dibuixen i també sobre quin paper juguem els catalans en aquesta revolució. En la qüestió catalana hem volgut saber si les nostres universitats són capaces de generar el talent necessari, com impulsar estratègicament el teixit empresarial en aquesta àrea, quin paper ha de tenir l’administració i què hauria de passar perquè la intel·ligència artificial també parli en català. En aquestes converses, els nostres interlocutors ens han transmès la urgència i la necessitat de treballar en la definició d’estratègies concretes de país per tal de no quedar socialment i estructuralment enrere en el que sembla una revolució sense límits, Artificial intelligence moves millions of euros and occupies an important part of governments’ political and strategic agendas. In this article we reflect on this varied concept through interviews with three people who hold important positions in the field of artificial intelligence and who also have in common the fact that they were born and trained in Catalonia. We talked with them about the origins of this technological revolution, its impact on social organization and the labour market, its ethical implications, the future trends which are emerging and also on what role we Catalans play in this. With regards to the Catalan question we wanted to know if our universities are capable of generating the necessary talent, how to strategically boost the fabric of business in this area, what role the Administration should play and what should happen for artificial intelligence to communicate in Catalan. Throughout these conversations, the interviewees have conveyed to us the urgency and need to work on the definition of country-specific strategies in order not to be left behind socially and structurally in what seems to be a revolution without limits., Aquest treball ha estat finançat en part pel Ministerio de Ciencia e Innovación mitjançant el programa Ramón i Cajal i per l’European Research Council (ERC) sota European Union’s Horizon 2020 research and innovation programme (grant agreement No 947657)., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Multilingual machine translation: Deep analysis of language-specific encoder-decoders

  • Escolano Peinado, Carlos
  • Ruiz Costa-Jussà, Marta
  • Rodríguez Fonollosa, José Adrián
State-of-the-art multilingual machine translation relies on a shared encoder-decoder. In this paper, we propose an alternative approach based on language-specific encoder-decoders, which can be easily extended to new languages by learning their corresponding modules. To establish a common interlingua representation, we simultaneously train N initial languages. Our experiments show that the proposed approach improves over the shared encoder-decoder for the initial languages and when adding new languages, without the need to retrain the remaining modules. All in all, our work closes the gap between shared and language-specific encoder-decoders, advancing toward modular multilingual machine translation systems that can be flexibly extended in lifelong learning settings., This work is supported by the European Research Council (ERC) under the European Union’sHorizon 2020 research and innovation programme (grant agreement No. 947657)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Converses al voltant de la intel·ligència artificial en clau catalana, Conversations around artificial intelligence from the catalan perspective

  • Ruiz Costa-Jussà, Marta
  • Melero Nogues, Maite
La intel·ligència artificial mou milions d’euros i ocupa una part important de les agendes polítiques i estratègiques dels governs. En aquest article reflexionem sobre aquest concepte difús i ho fem a través de tres entrevistes a tres persones que ocupen llocs rellevants en aquest àmbit, i que a més tenen en comú que són nascudes i formades a Catalunya. Amb elles hem parlat distesament dels orígens d’aquesta revolució tecnològica, del seu impacte sobre l’organització social i el mercat laboral, de les seves implicacions ètiques, de les tendències futures que es dibuixen i també sobre quin paper juguem els catalans en aquesta revolució. En la qüestió catalana hem volgut saber si les nostres universitats són capaces de generar el talent necessari, com impulsar estratègicament el teixit empresarial en aquesta àrea, quin paper ha de tenir l’administració i què hauria de passar perquè la intel·ligència artificial també parli en català. En aquestes converses, els nostres interlocutors ens han transmès la urgència i la necessitat de treballar en la definició d’estratègies concretes de país per tal de no quedar socialment i estructuralment enrere en el que sembla una revolució sense límits, Artificial intelligence moves millions of euros and occupies an important part of governments’ political and strategic agendas. In this article we reflect on this varied concept through interviews with three people who hold important positions in the field of artificial intelligence and who also have in common the fact that they were born and trained in Catalonia. We talked with them about the origins of this technological revolution, its impact on social organization and the labour market, its ethical implications, the future trends which are emerging and also on what role we Catalans play in this. With regards to the Catalan question we wanted to know if our universities are capable of generating the necessary talent, how to strategically boost the fabric of business in this area, what role the Administration should play and what should happen for artificial intelligence to communicate in Catalan. Throughout these conversations, the interviewees have conveyed to us the urgency and need to work on the definition of country-specific strategies in order not to be left behind socially and structurally in what seems to be a revolution without limits., Aquest treball ha estat finançat en part pel Ministerio de Ciencia e Innovación mitjançant el programa Ramón i Cajal i per l’European Research Council (ERC) sota European Union’s Horizon 2020 research and innovation programme (grant agreement No 947657)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Semantic and syntactic information for neural machine translation: Injecting features to the transformer

  • Armengol Estapé, Jordi
  • Ruiz Costa-Jussà, Marta
Introducing factors such as linguistic features has long been proposed in machine translation to improve the quality of translations. More recently, factored machine translation has proven to still be useful in the case of sequence-to-sequence systems. In this work, we investigate whether this gains hold in the case of the state-of-the-art architecture in neural machine translation, the Transformer, instead of recurrent architectures. We propose a new model, the Factored Transformer, to introduce an arbitrary number of word features in the source sequence in an attentional system. Specifically, we suggest two variants depending on the level at which the features are injected. Moreover, we suggest two combination mechanisms for the word features and words themselves. We experiment both with classical linguistic features and semantic features extracted from a linked data database, and with two low-resource datasets. With the best-found configuration, we show improvements of 0.8 BLEU over the baseline Transformer in the IWSLT German-to-English task. Moreover, we experiment with the more challenging FLoRes English-to-Nepali benchmark, which includes both low-resource and very distant languages, and obtain an improvement of 1.2 BLEU. These improvements are achieved with linguistic and not with semantic information., This work is supported by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant Agreement No. 947657)., Peer Reviewed, Postprint (published version)