Facebook has begun using neural network systems to process the billions of daily translations that occur on the backend of its platform. “Translations carried out with recurrent neural networks
(RNNs) were able to scale with the use of Caffe2, a deep learning framework open-sourced by Facebook in April,”
Venture Beat notes. “The switch also makes translations on Facebook
more likely to take into account things like slang, typos, and context.”
Read the whole story at Venture Beat »