Learning to Correspond Dynamical Systems

Nam Hee Kim, Zhaoming Xie, Michiel Panne
Proceedings of the 2nd Conference on Learning for Dynamics and Control, PMLR 120:105-117, 2020.

Abstract

Many dynamical systems exhibit similar structure, as often captured by hand-designed simplified models that can be used for analysis and control. We develop a method for learning to correspond pairs of dynamical systems via a learned latent dynamical system. Given trajectory data from two dynamical systems, we learn a shared latent state space and a shared latent dynamics model, along with an encoder-decoder pair for each of the original systems. With the learned correspondences in place, we can use a simulation of one system to produce an imagined motion of its counterpart. We can also simulate in the learned latent dynamics and synthesize the motions of both corresponding systems, as a form of bisimulation. We demonstrate the approach using pairs of controlled bipedal walkers, as well as by pairing a walker with a controlled pendulum.

Cite this Paper


BibTeX
@InProceedings{pmlr-v120-kim20a, title = {Learning to Correspond Dynamical Systems}, author = {Kim, Nam Hee and Xie, Zhaoming and van de Panne, Michiel}, booktitle = {Proceedings of the 2nd Conference on Learning for Dynamics and Control}, pages = {105--117}, year = {2020}, editor = {Bayen, Alexandre M. and Jadbabaie, Ali and Pappas, George and Parrilo, Pablo A. and Recht, Benjamin and Tomlin, Claire and Zeilinger, Melanie}, volume = {120}, series = {Proceedings of Machine Learning Research}, month = {10--11 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v120/kim20a/kim20a.pdf}, url = {https://proceedings.mlr.press/v120/kim20a.html}, abstract = {Many dynamical systems exhibit similar structure, as often captured by hand-designed simplified models that can be used for analysis and control. We develop a method for learning to correspond pairs of dynamical systems via a learned latent dynamical system. Given trajectory data from two dynamical systems, we learn a shared latent state space and a shared latent dynamics model, along with an encoder-decoder pair for each of the original systems. With the learned correspondences in place, we can use a simulation of one system to produce an imagined motion of its counterpart. We can also simulate in the learned latent dynamics and synthesize the motions of both corresponding systems, as a form of bisimulation. We demonstrate the approach using pairs of controlled bipedal walkers, as well as by pairing a walker with a controlled pendulum.} }
Endnote
%0 Conference Paper %T Learning to Correspond Dynamical Systems %A Nam Hee Kim %A Zhaoming Xie %A Michiel Panne %B Proceedings of the 2nd Conference on Learning for Dynamics and Control %C Proceedings of Machine Learning Research %D 2020 %E Alexandre M. Bayen %E Ali Jadbabaie %E George Pappas %E Pablo A. Parrilo %E Benjamin Recht %E Claire Tomlin %E Melanie Zeilinger %F pmlr-v120-kim20a %I PMLR %P 105--117 %U https://proceedings.mlr.press/v120/kim20a.html %V 120 %X Many dynamical systems exhibit similar structure, as often captured by hand-designed simplified models that can be used for analysis and control. We develop a method for learning to correspond pairs of dynamical systems via a learned latent dynamical system. Given trajectory data from two dynamical systems, we learn a shared latent state space and a shared latent dynamics model, along with an encoder-decoder pair for each of the original systems. With the learned correspondences in place, we can use a simulation of one system to produce an imagined motion of its counterpart. We can also simulate in the learned latent dynamics and synthesize the motions of both corresponding systems, as a form of bisimulation. We demonstrate the approach using pairs of controlled bipedal walkers, as well as by pairing a walker with a controlled pendulum.
APA
Kim, N.H., Xie, Z. & Panne, M.. (2020). Learning to Correspond Dynamical Systems. Proceedings of the 2nd Conference on Learning for Dynamics and Control, in Proceedings of Machine Learning Research 120:105-117 Available from https://proceedings.mlr.press/v120/kim20a.html.

Related Material