D
David Grangier
Researcher at Google
Publications - 108
Citations - 17040
David Grangier is an academic researcher from Google. The author has contributed to research in topics: Machine translation & Language model. The author has an hindex of 41, co-authored 103 publications receiving 12411 citations. Previous affiliations of David Grangier include Idiap Research Institute & Facebook.
Papers
More filters
Posted Content
fairseq: A Fast, Extensible Toolkit for Sequence Modeling.
Myle Ott,Sergey Edunov,Alexei Baevski,Angela Fan,Sam Gross,Nathan Ng,David Grangier,Michael Auli +7 more
TL;DR: fairseq as discussed by the authors is an open-source sequence modeling toolkit that allows researchers and developers to train custom models for translation, summarization, language modeling, and other text generation tasks, and supports distributed training across multiple GPUs and machines.
Proceedings Article
Convolutional Sequence to Sequence Learning
TL;DR: The authors introduced an architecture based entirely on convolutional neural networks, where computations over all elements can be fully parallelized during training and optimization is easier since the number of nonlinearities is fixed and independent of the input length.
Proceedings ArticleDOI
fairseq: A Fast, Extensible Toolkit for Sequence Modeling
Myle Ott,Sergey Edunov,Alexei Baevski,Angela Fan,Sam Gross,Nathan Ng,David Grangier,Michael Auli +7 more
TL;DR: Fairseq is an open-source sequence modeling toolkit that allows researchers and developers to train custom models for translation, summarization, language modeling, and other text generation tasks and supports distributed training across multiple GPUs and machines.
Proceedings Article
Language modeling with gated convolutional networks
TL;DR: A finite context approach through stacked convolutions, which can be more efficient since they allow parallelization over sequential tokens, is developed and is the first time a non-recurrent approach is competitive with strong recurrent models on these large scale language tasks.
Posted Content
Convolutional Sequence to Sequence Learning
TL;DR: The authors introduced an architecture based entirely on convolutional neural networks, where computations over all elements can be fully parallelized during training and optimization is easier since the number of nonlinearities is fixed and independent of the input length.