Learning Dynamical Systems with Side Information

Amir Ali Ahmadi, Bachir El Khadir
Proceedings of the 2nd Conference on Learning for Dynamics and Control, PMLR 120:718-727, 2020.

Abstract

We present a mathematical formalism and a computational framework for the problem of learning a dynamical system from noisy observations of a few trajectories and subject to side information (e.g., physical laws or contextual knowledge). We identify six classes of side information which can be imposed by semidefinite programming and that arise naturally in many applications. We demonstrate their value on two examples from epidemiology and physics. Some density results on polynomial dynamical systems that either exactly or approximately satisfy side information are also presented.

Cite this Paper

BibTeX
@InProceedings{pmlr-v120-ahmadi20a, title = {Learning Dynamical Systems with Side Information}, author = {Ahmadi, Amir Ali and Khadir, Bachir El}, booktitle = {Proceedings of the 2nd Conference on Learning for Dynamics and Control}, pages = {718--727}, 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/ahmadi20a/ahmadi20a.pdf}, url = {https://proceedings.mlr.press/v120/ahmadi20a.html}, abstract = { We present a mathematical formalism and a computational framework for the problem of learning a dynamical system from noisy observations of a few trajectories and subject to side information (e.g., physical laws or contextual knowledge). We identify six classes of side information which can be imposed by semidefinite programming and that arise naturally in many applications. We demonstrate their value on two examples from epidemiology and physics. Some density results on polynomial dynamical systems that either exactly or approximately satisfy side information are also presented. } }
Endnote
%0 Conference Paper %T Learning Dynamical Systems with Side Information %A Amir Ali Ahmadi %A Bachir El Khadir %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-ahmadi20a %I PMLR %P 718--727 %U https://proceedings.mlr.press/v120/ahmadi20a.html %V 120 %X We present a mathematical formalism and a computational framework for the problem of learning a dynamical system from noisy observations of a few trajectories and subject to side information (e.g., physical laws or contextual knowledge). We identify six classes of side information which can be imposed by semidefinite programming and that arise naturally in many applications. We demonstrate their value on two examples from epidemiology and physics. Some density results on polynomial dynamical systems that either exactly or approximately satisfy side information are also presented.
APA
Ahmadi, A.A. & Khadir, B.E.. (2020). Learning Dynamical Systems with Side Information. Proceedings of the 2nd Conference on Learning for Dynamics and Control, in Proceedings of Machine Learning Research 120:718-727 Available from https://proceedings.mlr.press/v120/ahmadi20a.html.

Related Material