Regret Bounds for Adaptive Nonlinear Control

Nicholas M. Boffi, Stephen Tu, Jean-Jacques E. Slotine
Proceedings of the 3rd Conference on Learning for Dynamics and Control, PMLR 144:471-483, 2021.

Abstract

We study the problem of adaptively controlling a known discrete-time nonlinear system subject to unmodeled disturbances. We prove the first finite-time regret bounds for adaptive nonlinear control with matched uncertainty in the stochastic setting, showing that the regret suffered by certainty equivalence adaptive control, compared to an oracle controller with perfect knowledge of the un-modeled disturbances, is upper bounded by $\widetilde{O}(\sqrt{T})$ in expectation. Furthermore, we show that when the input is subject to a k timestep delay, the regret degrades to $\widetilde{O}(k\sqrt{T})$. Our analysis draws connections between classical stability notions in nonlinear control theory (Lyapunov stability and contraction theory) and modern regret analysis from online convex optimization. The use of stability theory allows us to analyze the challenging infinite-horizon single trajectory setting.

Cite this Paper


BibTeX
@InProceedings{pmlr-v144-boffi21a, title = {Regret Bounds for Adaptive Nonlinear Control}, author = {Boffi, Nicholas M. and Tu, Stephen and Slotine, Jean-Jacques E.}, booktitle = {Proceedings of the 3rd Conference on Learning for Dynamics and Control}, pages = {471--483}, year = {2021}, editor = {Jadbabaie, Ali and Lygeros, John and Pappas, George J. and A. Parrilo, Pablo and Recht, Benjamin and Tomlin, Claire J. and Zeilinger, Melanie N.}, volume = {144}, series = {Proceedings of Machine Learning Research}, month = {07 -- 08 June}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v144/boffi21a/boffi21a.pdf}, url = {https://proceedings.mlr.press/v144/boffi21a.html}, abstract = {We study the problem of adaptively controlling a known discrete-time nonlinear system subject to unmodeled disturbances. We prove the first finite-time regret bounds for adaptive nonlinear control with matched uncertainty in the stochastic setting, showing that the regret suffered by certainty equivalence adaptive control, compared to an oracle controller with perfect knowledge of the un-modeled disturbances, is upper bounded by $\widetilde{O}(\sqrt{T})$ in expectation. Furthermore, we show that when the input is subject to a k timestep delay, the regret degrades to $\widetilde{O}(k\sqrt{T})$. Our analysis draws connections between classical stability notions in nonlinear control theory (Lyapunov stability and contraction theory) and modern regret analysis from online convex optimization. The use of stability theory allows us to analyze the challenging infinite-horizon single trajectory setting.} }
Endnote
%0 Conference Paper %T Regret Bounds for Adaptive Nonlinear Control %A Nicholas M. Boffi %A Stephen Tu %A Jean-Jacques E. Slotine %B Proceedings of the 3rd Conference on Learning for Dynamics and Control %C Proceedings of Machine Learning Research %D 2021 %E Ali Jadbabaie %E John Lygeros %E George J. Pappas %E Pablo A. Parrilo %E Benjamin Recht %E Claire J. Tomlin %E Melanie N. Zeilinger %F pmlr-v144-boffi21a %I PMLR %P 471--483 %U https://proceedings.mlr.press/v144/boffi21a.html %V 144 %X We study the problem of adaptively controlling a known discrete-time nonlinear system subject to unmodeled disturbances. We prove the first finite-time regret bounds for adaptive nonlinear control with matched uncertainty in the stochastic setting, showing that the regret suffered by certainty equivalence adaptive control, compared to an oracle controller with perfect knowledge of the un-modeled disturbances, is upper bounded by $\widetilde{O}(\sqrt{T})$ in expectation. Furthermore, we show that when the input is subject to a k timestep delay, the regret degrades to $\widetilde{O}(k\sqrt{T})$. Our analysis draws connections between classical stability notions in nonlinear control theory (Lyapunov stability and contraction theory) and modern regret analysis from online convex optimization. The use of stability theory allows us to analyze the challenging infinite-horizon single trajectory setting.
APA
Boffi, N.M., Tu, S. & Slotine, J.E.. (2021). Regret Bounds for Adaptive Nonlinear Control. Proceedings of the 3rd Conference on Learning for Dynamics and Control, in Proceedings of Machine Learning Research 144:471-483 Available from https://proceedings.mlr.press/v144/boffi21a.html.

Related Material