Sequence to Sequence Learning with Neural Networks

NeurIPS 2014 Ilya SutskeverOriol VinyalsQuoc 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... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Traffic Prediction PeMS-M FC-LSTM MAE (60 min) 4.16 # 5
Machine Translation WMT2014 English-French SMT+LSTM5 BLEU score 36.5 # 22
Machine Translation WMT2014 English-French LSTM BLEU score 34.8 # 26

Methods used in the Paper