Viet-Anh on Software Logo

What is: Variational Autoencoder?

SourceAuto-Encoding Variational Bayes
Year2000
Data SourceCC BY-SA - https://paperswithcode.com

A Variational Autoencoder is a type of likelihood-based generative model. It consists of an encoder, that takes in data xx as input and transforms this into a latent representation zz, and a decoder, that takes a latent representation zz and returns a reconstruction x^\hat{x}. Inference is performed via variational inference to approximate the posterior of the model.