On the Properties of Neural Machine Translation: Encoder–Decoder Approaches
Kyunghyun Cho, Bart van Merriënboer, Dzmitry Bahdanau, Yoshua Bengio
Neural machine translation is a relatively new approach to statistical machine translation based purely on neural networks. The neural machine translation models often consist of an encoder and a decoder. The encoder extracts a fixed-length representation from a variable-length input sentence, and the decoder generates a correct translation from this representation. In this paper, we focus on analyzing the properties of the neural machine translation using two models; RNN Encoder-Decoder and a newly proposed gated recursive convolutional neural network. We show that the neural machine translation performs relatively well on short sentences without unknown words, but its performance degrades rapidly as the length of the sentence and the number of unknown words increase. Furthermore, we find that the proposed gated recursive convolutional network learns a grammatical structure of a sentence automatically.
Citation edge evidence
- Deep learning for healthcare: review, opportunities and challenges→On the Properties of Neural Machine Translation: Encoder–Decoder ApproachesLegacy edge — source detail unavailable
- End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF→On the Properties of Neural Machine Translation: Encoder–Decoder ApproachesLegacy edge — source detail unavailable
- Supervised Learning of Universal Sentence Representations from Natural\n Language Inferen…→On the Properties of Neural Machine Translation: Encoder–Decoder ApproachesLegacy edge — source detail unavailable
- A Comprehensive Survey of Graph Embedding: Problems, Techniques, and Applications→On the Properties of Neural Machine Translation: Encoder–Decoder ApproachesLegacy edge — source detail unavailable
- Natural language processing: state of the art, current trends and challenges→On the Properties of Neural Machine Translation: Encoder–Decoder ApproachesLegacy edge — source detail unavailable
- Counterfactual Multi-Agent Policy Gradients→On the Properties of Neural Machine Translation: Encoder–Decoder ApproachesLegacy edge — source detail unavailable
- On the Properties of Neural Machine Translation: Encoder–Decoder Approaches→Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Transla…Legacy edge — source detail unavailable
- On the Properties of Neural Machine Translation: Encoder–Decoder Approaches→Sequence to Sequence Learning with Neural NetworksLegacy edge — source detail unavailable
- On the Properties of Neural Machine Translation: Encoder–Decoder Approaches→ADADELTA: An Adaptive Learning Rate MethodLegacy edge — source detail unavailable
- On the Properties of Neural Machine Translation: Encoder–Decoder Approaches→Statistical phrase-based translationLegacy edge — source detail unavailable
- Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Transla…→Statistical phrase-based translationLegacy edge — source detail unavailable
- Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Transla…→ADADELTA: An Adaptive Learning Rate MethodLegacy edge — source detail unavailable
- Deep learning for healthcare: review, opportunities and challenges→Sequence to Sequence Learning with Neural NetworksLegacy edge — source detail unavailable
- End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF→ADADELTA: An Adaptive Learning Rate MethodLegacy edge — source detail unavailable
- Supervised Learning of Universal Sentence Representations from Natural\n Language Inferen…→Sequence to Sequence Learning with Neural NetworksLegacy edge — source detail unavailable
What this paper cites, inside the corpus
What cites it, inside the corpus
Links
Download BibTeX · Download RIS · DOI · Open access copy · OpenAlex record
Topics
| Natural Language Processing Techniques | Computer Science |
| Topic Modeling | Computer Science |
| Machine Learning in Bioinformatics | Biochemistry, Genetics and Molecular Biology |
Is this record sound?
complete
Nothing in this record contradicts itself and no field we check is missing.
- supports4 author record(s) attached.
- supports12 reference(s) recorded.
- neutralThe DOI carries no year to check against.
- supportsA title is present.
Provenance
sha256 5cad55ac5d4d41e1…