NeurIPS 2015poster1003 citations

Embed to Control: A Locally Linear Latent Dynamics Model for Control from Raw Images

Manuel Watter, Jost Springenberg, Joschka Boedecker, Martin Riedmiller

Abstract

We introduce Embed to Control (E2C), a method for model learning and control of non-linear dynamical systems from raw pixel images. E2C consists of a deep generative model, belonging to the family of variational autoencoders, that learns to generate image trajectories from a latent space in which the dynamics is constrained to be locally linear. Our model is derived directly from an optimal control formulation in latent space, supports long-term prediction of image sequences and exhibits strong performance on a variety of complex control problems.

BibTeX
@inproceedings{NIPS2015_a1afc58c,
 author = {Watter, Manuel and Springenberg, Jost and Boedecker, Joschka and Riedmiller, Martin},
 booktitle = {Advances in Neural Information Processing Systems},
 editor = {C. Cortes and N. Lawrence and D. Lee and M. Sugiyama and R. Garnett},
 pages = {},
 publisher = {Curran Associates, Inc.},
 title = {Embed to Control: A Locally Linear Latent Dynamics Model for Control from Raw Images},
 url = {https://proceedings.neurips.cc/paper_files/paper/2015/file/a1afc58c6ca9540d057299ec3016d726-Paper.pdf},
 volume = {28},
 year = {2015}
}
Embed to Control: A Locally Linear Latent Dynamics Model for Control from Raw Images · NeurIPS 2015