Dzmitry Bahdanau

E899028

computer scientist researcher scientific paper

Dzmitry Bahdanau is a computer scientist best known for pioneering the neural attention mechanism in sequence-to-sequence models, which transformed neural machine translation and modern deep learning.

Try in SPARQL Jump to: Surface forms Statements Referenced by

Observed surface forms (1)

Surface form	Occurrences
Neural Machine Translation by Jointly Learning to Align and Translate	0

Statements (44)

Predicate	Object
instanceOf	computer scientist ⓘ researcher ⓘ scientific paper ⓘ
advisor	Yoshua Bengio NERFINISHED ⓘ
alsoKnownAs	Dmitry Bahdanau NERFINISHED ⓘ
associatedWith	Université de Montréal NERFINISHED ⓘ
author	Dzmitry Bahdanau NERFINISHED ⓘ Kyunghyun Cho NERFINISHED ⓘ Yoshua Bengio NERFINISHED ⓘ
citationsCountRange	>10000 citations for attention NMT paper ⓘ
coAuthor	Kyunghyun Cho NERFINISHED ⓘ Yoshua Bengio NERFINISHED ⓘ
countryOfCitizenship	Belarus NERFINISHED ⓘ
fieldOfWork	deep learning ⓘ machine learning ⓘ natural language processing ⓘ neural machine translation ⓘ neural machine translation ⓘ
gender	male ⓘ
hasAcademicAdvisor	Yoshua Bengio NERFINISHED ⓘ
hasContribution	demonstrated improvements over encoder-decoder models without attention ⓘ enabled alignment learning jointly with translation in NMT ⓘ introduced additive attention mechanism for sequence-to-sequence models ⓘ popularized attention mechanisms in NLP research ⓘ
influenced	development of attention mechanisms in deep learning ⓘ modern neural machine translation systems ⓘ transformer-based neural network architectures ⓘ
influencedBy	Kyunghyun Cho NERFINISHED ⓘ Yoshua Bengio NERFINISHED ⓘ
knownFor	neural attention mechanism in sequence-to-sequence models ⓘ pioneering attention-based neural machine translation ⓘ
languageOfWorkOrName	Belarusian ⓘ English ⓘ Russian ⓘ
mainSubject	attention mechanism ⓘ
nativeLanguage	Belarusian ⓘ
notableIdea	additive attention NERFINISHED ⓘ soft alignment in neural translation ⓘ
notableWork	Neural Machine Translation by Jointly Learning to Align and Translate NERFINISHED ⓘ
publicationYear	2014 ⓘ
researchInterest	neural network architectures ⓘ representation learning ⓘ sequence-to-sequence learning ⓘ
workLocation	Montreal NERFINISHED ⓘ

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.

Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation → author → Dzmitry Bahdanau ⓘ