Sequence to Sequence Learning with Neural Networks

NeurIPS 2014  ·  Ilya Sutskever, Oriol Vinyals, Quoc V. Le ·

Deep Neural Networks (DNNs) are powerful models that have achieved excellent performance on difficult learning tasks. Although DNNs work well whenever large labeled training sets are available, they cannot be used to map sequences to sequences. In this paper, we present a general end-to-end approach to sequence learning that makes minimal assumptions on the sequence structure. Our method uses a multilayered Long Short-Term Memory (LSTM) to map the input sequence to a vector of a fixed dimensionality, and then another deep LSTM to decode the target sequence from the vector. Our main result is that on an English to French translation task from the WMT'14 dataset, the translations produced by the LSTM achieve a BLEU score of 34.8 on the entire test set, where the LSTM's BLEU score was penalized on out-of-vocabulary words. Additionally, the LSTM did not have difficulty on long sentences. For comparison, a phrase-based SMT system achieves a BLEU score of 33.3 on the same dataset. When we used the LSTM to rerank the 1000 hypotheses produced by the aforementioned SMT system, its BLEU score increases to 36.5, which is close to the previous best result on this task. The LSTM also learned sensible phrase and sentence representations that are sensitive to word order and are relatively invariant to the active and the passive voice. Finally, we found that reversing the order of the words in all source sentences (but not target sentences) improved the LSTM's performance markedly, because doing so introduced many short term dependencies between the source and the target sentence which made the optimization problem easier.

PDF Abstract NeurIPS 2014 PDF NeurIPS 2014 Abstract


Results from the Paper

Ranked #5 on Traffic Prediction on PeMS-M (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
Time Series Forecasting PeMSD7 FC-LSTM 9 steps MAE 4.16 # 5
Traffic Prediction PeMS-M FC-LSTM MAE (60 min) 4.16 # 5
Machine Translation WMT2014 English-French SMT+LSTM5 BLEU score 36.5 # 42
Machine Translation WMT2014 English-French LSTM BLEU score 34.8 # 46