A Deep Autoencoder
(Ruslan Salakhutdinov)
28x28
      1000  neurons
They always looked like a
really nice way to do non-
linear dimensionality
reduction:
But it is very difficult to
optimize deep
autoencoders using
backpropagation.
We now have a much better
way to optimize them.
500 neurons
250 neurons
linear
units
30
250 neurons
500 neurons
      1000  neurons
28x28