Learning Disentangled Representations with Semi-Supervised Deep Generative Models

Understand

Variational autoencoders (VAEs) learn representations of data by jointly training a probabilistic encoder and decoder network.

• Typically these models encode all features of the data into a single variable.
• Here we are interested in learning disentangled representations that encode distinct aspects of the data into separate variables.
• We propose to learn such representations using model architectures that generalise from standard VAEs, employing a general graphical model structure in the encoder and decoder.