Skip to content
/ vae_ad Public

Repository containing vanilla variational autoencoder for anomaly detection

License

Notifications You must be signed in to change notification settings

edhenry/vae_ad

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

vae_ad

Repository containing vanilla variational autoencoder for anomaly detection.

Autoencoding Variational Bayes

This repository is an extension of the work done by Kingma et al. in Autoencoding Variational Bayes

The idea is to use this approach as a way to model, probabilistically, that variational autoencoders can be used for novelty detection given that the encoder/decoder architecture is finding a latent set of codes (z) that are representative of the underlying data generating distribution. This repository is using MNIST as an example, but the concept still applies to many other domains that have a continuous representations as input.

About

Repository containing vanilla variational autoencoder for anomaly detection

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages