Back to Newsroom

A disentangled recognition and nonlinear dynamics model for Unsupervised Learning

Dive into the conversation around unsupervised learning and its impact within Raffle to understand our unique structure.
Back to Blog

Authors: Marco Fraccaro, Simon Kamronn, Ulrich Paquet, Ole Winther

Publication date: 2017

Conference: Advances in Neural Information Processing Systems

We introduce the Kalman variational auto-encoder, a framework for unsupervised learning of sequential data that disentangles two latent representations: an object's representation, coming from a recognition model, and a latent state describing its dynamics.

As a result, the evolution of the world can be imagined and missing data imputed, both without the need to generate high dimensional frames at each time step. The model is trained end-to-end on videos of a variety of simulated physical systems, and outperforms competing methods in generative and missing data imputation tasks.