NeurIPS 2019oral133 citations
Logarithmic Regret for Online Control
Naman Agarwal, Elad Hazan, Karan Singh
Abstract
We study optimal regret bounds for control in linear dynamical systems under adversarially changing strongly convex cost functions, given the knowledge of transition dynamics. This includes several well studied and influential frameworks such as the Kalman filter and the linear quadratic regulator. State of the art methods achieve regret which scales as T^0.5, where T is the time horizon.
BibTeX
@inproceedings{NEURIPS2019_78719f11,
author = {Agarwal, Naman and Hazan, Elad and Singh, Karan},
booktitle = {Advances in Neural Information Processing Systems},
editor = {H. Wallach and H. Larochelle and A. Beygelzimer and F. d\textquotesingle Alch\'{e}-Buc and E. Fox and R. Garnett},
pages = {},
publisher = {Curran Associates, Inc.},
title = {Logarithmic Regret for Online Control},
url = {https://proceedings.neurips.cc/paper_files/paper/2019/file/78719f11fa2df9917de3110133506521-Paper.pdf},
volume = {32},
year = {2019}
}