The Nonstochastic Control Problem
Abstract
We consider the problem of controlling an unknown linear dynamical system in the presence of (nonstochastic) adversarial perturbations and adversarial convex loss functions. In contrast to classical control, the a priori determination of an optimal controller here is hindered by the latter’s dependence on the yet unknown perturbations and costs. Instead, we measure regret against an optimal linear policy in hindsight, and give the first efficient algorithm that guarantees a sublinear regret bound, scaling as T^(2/3), in this setting.
Cite
Text
Hazan et al. "The Nonstochastic Control Problem." Proceedings of the 31st International Conference on Algorithmic Learning Theory, 2020.Markdown
[Hazan et al. "The Nonstochastic Control Problem." Proceedings of the 31st International Conference on Algorithmic Learning Theory, 2020.](https://mlanthology.org/alt/2020/hazan2020alt-nonstochastic/)BibTeX
@inproceedings{hazan2020alt-nonstochastic,
title = {{The Nonstochastic Control Problem}},
author = {Hazan, Elad and Kakade, Sham and Singh, Karan},
booktitle = {Proceedings of the 31st International Conference on Algorithmic Learning Theory},
year = {2020},
pages = {408-421},
volume = {117},
url = {https://mlanthology.org/alt/2020/hazan2020alt-nonstochastic/}
}