dzmitry bahdanau, kyunghyun cho, and yoshua bengio. 2014. neural machine translation by jointly learning to align and
Neural machine translation is a recently proposed approach to machine translation. Unlike the traditional statistical machine translation, the neural machine translation aims at building a single neural network that can be jointly tuned to maximize the translation performance.
The models proposed as of late for brain machine interpretation frequently have a place with a group of encoder-decoders and comprises of an encoder that encodes a source sentence into a fixed-length vector from which a decoder creates an interpretation.
In this paper, we guess that the utilization of a fixed-length vector is a bottleneck in working on the exhibition of this essential encoder-decoder engineering, and propose to broaden this by permitting a model to naturally (delicate )look for parts of a source sentence that are pertinent to anticipating an objective word, without shaping these parts as a hard section unequivocally.
With this new methodology, we accomplish an interpretation execution equivalent to the current cutting edge state put together framework with respect to the undertaking of English-to-French interpretation. Moreover, subjective examination uncovers that the (delicate )arrangements found by the model concur well with our instinct.
to learn more about neural machine translation
brainly.com/question/3299445
#SPJ4
