Now that we've learned about the components of RNNs, let's dive into what we can do with them. In this section, we'll look at two primary examples: machine translation and generating image captions. Later on in this book, we'll utilize RNNs to build a variety of end-to-end systems.
Sequence processing with RNNs
Neural machine translation
Machine translation represents a sequence-to-sequence problem; you'll frequently see these networks described as sequence-to-sequence (or Seq2Seq) models. Instead of utilizing traditional techniques that involve feature engineering and n-gram counts, neural machine translation maps the overall meaning of a sentence to a singular vector, and we then generate a translation...
