Open AccessProceedings Article
Neural Machine Translation by Jointly Learning to Align and Translate
TLDR
It is conjecture that the use of a fixed-length vector is a bottleneck in improving the performance of this basic encoder-decoder architecture, and it is proposed to extend this by allowing a model to automatically (soft-)search for parts of a source sentence that are relevant to predicting a target word, without having to form these parts as a hard segment explicitly.Abstract:
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 recently for neural machine translation often belong to a family of encoder-decoders and consists of an encoder that encodes a source sentence into a fixed-length vector from which a decoder generates a translation. In this paper, we conjecture that the use of a fixed-length vector is a bottleneck in improving the performance of this basic encoder-decoder architecture, and propose to extend this by allowing a model to automatically (soft-)search for parts of a source sentence that are relevant to predicting a target word, without having to form these parts as a hard segment explicitly. With this new approach, we achieve a translation performance comparable to the existing state-of-the-art phrase-based system on the task of English-to-French translation. Furthermore, qualitative analysis reveals that the (soft-)alignments found by the model agree well with our intuition.read more
Citations
More filters
Posted Content
Neural AMR: Sequence-to-Sequence Models for Parsing and Generation
TL;DR: This work presents a novel training procedure that can lift the limitation of the relatively limited amount of labeled data and the non-sequential nature of the AMR graphs, and presents strong evidence that sequence-based AMR models are robust against ordering variations of graph-to-sequence conversions.
Journal ArticleDOI
A domain-specific supercomputer for training deep neural networks
Norman P. Jouppi,Doe Hyun Yoon,George Kurian,Sheng Li,Nishant Patil,James Laudon,Cliff Young,David A. Patterson +7 more
TL;DR: Google's TPU supercomputers train deep neural networks 50x faster than general-purpose supercomputing running a high-performance computing benchmark.
Proceedings ArticleDOI
What Would You Expect? Anticipating Egocentric Actions With Rolling-Unrolling LSTMs and Modality Attention
TL;DR: Rulstm as mentioned in this paper proposes an architecture able to anticipate actions at multiple temporal scales using two LSTMs to summarize the past and formulate predictions about the future, which is ranked first in the public leaderboard of the EPIC-Kitchens egocentric action anticipation challenge 2019.
Proceedings ArticleDOI
A Time-Restricted Self-Attention Layer for ASR
TL;DR: This paper applies a restricted self-attention mechanism (with multiple heads) to speech recognition, and tries introducing attention layers into TDNN architectures, and replacing LSTM layers with attention layers in TDNN+LSTM architectures.
Proceedings ArticleDOI
Classical Structured Prediction Losses for Sequence to Sequence Learning.
TL;DR: The authors survey a range of classical objective functions that have been widely used to train linear models for structured prediction and apply them to neural sequence-to-sequence models, showing that these losses can perform surprisingly well by slightly outperforming beam search optimization in a like for like setup.
References
More filters
Journal ArticleDOI
Long short-term memory
TL;DR: A novel, efficient, gradient based method called long short-term memory (LSTM) is introduced, which can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units.
Proceedings ArticleDOI
Learning Phrase Representations using RNN Encoder--Decoder for Statistical Machine Translation
Kyunghyun Cho,Bart van Merriënboer,Caglar Gulcehre,Dzmitry Bahdanau,Fethi Bougares,Holger Schwenk,Yoshua Bengio,Yoshua Bengio,Yoshua Bengio +8 more
TL;DR: In this paper, the encoder and decoder of the RNN Encoder-Decoder model are jointly trained to maximize the conditional probability of a target sequence given a source sequence.
Journal ArticleDOI
Learning long-term dependencies with gradient descent is difficult
TL;DR: This work shows why gradient based learning algorithms face an increasingly difficult problem as the duration of the dependencies to be captured increases, and exposes a trade-off between efficient learning by gradient descent and latching on information for long periods.
Journal ArticleDOI
Bidirectional recurrent neural networks
Mike Schuster,Kuldip K. Paliwal +1 more
TL;DR: It is shown how the proposed bidirectional structure can be easily modified to allow efficient estimation of the conditional posterior probability of complete symbol sequences without making any explicit assumption about the shape of the distribution.
Journal ArticleDOI
A neural probabilistic language model
TL;DR: The authors propose to learn a distributed representation for words which allows each training sentence to inform the model about an exponential number of semantically neighboring sentences, which can be expressed in terms of these representations.