University of Cambridge > Talks.cam > Machine Learning @ CUED > Variational autoencoders with latent graphical models

Variational autoencoders with latent graphical models

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Zoubin Ghahramani.

We propose a general modeling and inference framework that composes probabilistic graphical models with deep learning methods, in a way that combines their respective strengths. Our model family combines graphical structure in latent variables with neural network observation models. For inference, we use variational recognition networks to produce local evidence summaries, and combine them using exact graphical model inference. We illustrate this framework with several example models, and an application to automatic mouse behavior modeling from video.

This talk is part of the Machine Learning @ CUED series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2019 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity