Advanced Deep Learning with Keras

Variational Autoencoders
(VAEs)
Similar to Generative Adversarial Networks (GANs) that we've discussed in
the previous chapters, Variational Autoencoders (VAEs) [1] belong to the family
of generative models. The generator of VAE is able to produce meaningful outputs
while navigating its continuous latent space. The possible attributes of the decoder
outputs are explored through the latent vector.
In GANs, the focus is on how to arrive at a model that approximates the input
distribution. VAEs attempt to model the input distribution from a decodable
continuous latent space. This is one of the possible underlying reasons why
GANs are able to generate more realistic signals when compared to VAEs. For
example, in image generation, GANs are able to produce more realistic looking
images while VAEs in comparison generate images that are less sharp.
Within VAEs, the focus is on the variational inference of latent codes.
Therefore, VAEs provide a suitable framework for both learning and efficient
Bayesian inference with latent variables. For example, VAEs with disentangled
representations enable latent code reuse for transfer learning.
In terms of structure, VAEs bear a resemblance to an autoencoder. They are
also made up of an encoder (also known as recognition or inference model)
and a decoder (also known as a generative model). Both VAEs and autoencoders
attempt to reconstruct the input data while learning the latent vector. However,
unlike autoencoders, the latent space of VAEs is continuous, and the decoder itself
is used as a generative model.
[ 237 ]