Skip to content

Instantly share code, notes, and snippets.

@jxnl
Last active March 28, 2017 17:54
Show Gist options
  • Save jxnl/4ec20dcf054b798cd3b80764213d9f60 to your computer and use it in GitHub Desktop.
Save jxnl/4ec20dcf054b798cd3b80764213d9f60 to your computer and use it in GitHub Desktop.

motivation

Variational Autoencoders (VAE) are extremely appealing as they allow for learning complicated distributions taking advantage of recent progress in gradient descent algorithms and accelerated processing with GPUs. The latent space of regular autoencoders are typically very sparse and unrestricted, making it difficult to generate data robust to variations of the latent variables. Traditional VAEs use normal priors regularize the latent space however these assuptions are unsound when it comes to complex data such as images or text.

challenge

We propose Dirichlet priors for a multinomial latent space. This latent space allows us to explore the data by interpreting $P(\theta_i)$ as the likelihood of belonging to a latent class, as a form of non-linear soft clustering.

contributions

We explore various methods for regularizing an Autoencoder so that the latent space is multinomial, and compare how these methods perform on a classification task using the latent space as the input to a k-nearest neighbours algorithm.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment