- -

NMT-Keras: a Very Flexible Toolkit with a Focus on Interactive NMT and Online Learning

RiuNet: Repositorio Institucional de la Universidad Politécnica de Valencia

Compartir/Enviar a

Citas

Estadísticas

  • Estadisticas de Uso

NMT-Keras: a Very Flexible Toolkit with a Focus on Interactive NMT and Online Learning

Mostrar el registro sencillo del ítem

Ficheros en el ítem

dc.contributor.author Peris-Abril, Álvaro es_ES
dc.contributor.author Casacuberta Nolla, Francisco es_ES
dc.date.accessioned 2020-06-03T05:53:18Z
dc.date.available 2020-06-03T05:53:18Z
dc.date.issued 2018-09-10 es_ES
dc.identifier.issn 0032-6585 es_ES
dc.identifier.uri http://hdl.handle.net/10251/145118
dc.description.abstract [EN] We present NMT-Keras, a flexible toolkit for training deep learning models, which puts a particular emphasis on the development of advanced applications of neural machine translation systems, such as interactive-predictive translation protocols and long-term adaptation of the translation system via continuous learning. NMT-Keras is based on an extended version of the popular Keras library, and it runs on Theano and TensorFlow. State-of-the-art neural machine translation models are deployed and used following the high-level framework provided by Keras. Given its high modularity and flexibility, it also has been extended to tackle different problems, such as image and video captioning, sentence classification and visual question answering. es_ES
dc.description.sponsorship Much of our Keras fork and the Multimodal Keras Wrapper libraries were developed together with Marc Bolaños. We also acknowledge the rest of contributors to these open-source projects. The research leading this work received funding from grants PROMETEO/2018/004 and CoMUN-HaT - TIN2015-70924-C2-1-R. We finally acknowledge NVIDIA Corporation for the donation of GPUs used in this work. es_ES
dc.language Inglés es_ES
dc.publisher Univerzita Karlova es_ES
dc.relation.ispartof The Prague Bulletin of Mathematical Linguistics es_ES
dc.rights Reconocimiento - No comercial - Sin obra derivada (by-nc-nd) es_ES
dc.subject Neural Machine Translation es_ES
dc.subject.classification LENGUAJES Y SISTEMAS INFORMATICOS es_ES
dc.title NMT-Keras: a Very Flexible Toolkit with a Focus on Interactive NMT and Online Learning es_ES
dc.type Artículo es_ES
dc.identifier.doi 10.2478/pralin-2018-0010 es_ES
dc.relation.projectID info:eu-repo/grantAgreement/GVA//PROMETEO%2F2018%2F004/ es_ES
dc.relation.projectID info:eu-repo/grantAgreement/MINECO//TIN2015-70924-C2-1-R/ES/CONTEXTO, MULTIMODALIDAD Y COLABORACION DEL USUARIO EN PROCESADO DE TEXTO MANUSCRITO/ es_ES
dc.rights.accessRights Abierto es_ES
dc.contributor.affiliation Universitat Politècnica de València. Departamento de Sistemas Informáticos y Computación - Departament de Sistemes Informàtics i Computació es_ES
dc.description.bibliographicCitation Peris-Abril, Á.; Casacuberta Nolla, F. (2018). NMT-Keras: a Very Flexible Toolkit with a Focus on Interactive NMT and Online Learning. The Prague Bulletin of Mathematical Linguistics. 111:113-124. https://doi.org/10.2478/pralin-2018-0010 es_ES
dc.description.accrualMethod S es_ES
dc.relation.publisherversion https://doi.org/10.2478/pralin-2018-0010 es_ES
dc.description.upvformatpinicio 113 es_ES
dc.description.upvformatpfin 124 es_ES
dc.type.version info:eu-repo/semantics/publishedVersion es_ES
dc.description.volume 111 es_ES
dc.relation.pasarela S\368399 es_ES
dc.contributor.funder Generalitat Valenciana es_ES
dc.contributor.funder Ministerio de Economía y Competitividad es_ES


Este ítem aparece en la(s) siguiente(s) colección(ones)

Mostrar el registro sencillo del ítem