Rnn
Notation
- one hot encoding
Why not a standard network?
Problems:
- Inputs, outputs can be different lengths in different examples.
- Doesn’t share features learned across different positions of text.
Recurrent Neural Networks
- Forward Propagation
Backpropagagtion through time
- many to one (Sentiment classification)
- many to many (machine translation)
- one to many (music generation)