DeePC-Hunt: Data-enabled Predictive Control Hyperparameter Tuning via Differentiable Optimization

Michael Cummins, Alberto Padoan, Keith Moffat, Florian Dorfler, John Lygeros
Proceedings of the 7th Annual Learning for Dynamics \& Control Conference, PMLR 283:673-685, 2025.

Abstract

This paper introduces Data-enabled Predictive Control Hyperparameter Tuning via Differentiable Optimization (DeePC-Hunt), a backpropagation-based method for automatic hyperparameter tuning of the DeePC algorithm. The necessity for such a method arises from the importance of hyperparameter selection to achieve satisfactory closed-loop DeePC performance. The standard methods for hyperparameter selection are to either optimize the open-loop performance, or use manual guess-and-check. Optimizing the open-loop performance can result in unacceptable closed-loop behavior, while manual guess-and-check can pose safety challenges. DeePC-Hunt provides an alternative method for hyperparameter tuning which uses an approximate model of the system dynamics and backpropagation to directly optimize hyperparameters for the closed-loop DeePC performance. Numerical simulations demonstrate the effectiveness of DeePC in combination with DeePC-Hunt in a complex stabilization task for a nonlinear system and its superiority over model-based control strategies in terms of robustness to model misspecifications.

Cite this Paper

BibTeX
@InProceedings{pmlr-v283-cummins25b, title = {DeePC-Hunt: Data-enabled Predictive Control Hyperparameter Tuning via Differentiable Optimization}, author = {Cummins, Michael and Padoan, Alberto and Moffat, Keith and Dorfler, Florian and Lygeros, John}, booktitle = {Proceedings of the 7th Annual Learning for Dynamics \& Control Conference}, pages = {673--685}, year = {2025}, editor = {Ozay, Necmiye and Balzano, Laura and Panagou, Dimitra and Abate, Alessandro}, volume = {283}, series = {Proceedings of Machine Learning Research}, month = {04--06 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v283/main/assets/cummins25b/cummins25b.pdf}, url = {https://proceedings.mlr.press/v283/cummins25b.html}, abstract = {This paper introduces Data-enabled Predictive Control Hyperparameter Tuning via Differentiable Optimization (DeePC-Hunt), a backpropagation-based method for automatic hyperparameter tuning of the DeePC algorithm. The necessity for such a method arises from the importance of hyperparameter selection to achieve satisfactory closed-loop DeePC performance. The standard methods for hyperparameter selection are to either optimize the open-loop performance, or use manual guess-and-check. Optimizing the open-loop performance can result in unacceptable closed-loop behavior, while manual guess-and-check can pose safety challenges. DeePC-Hunt provides an alternative method for hyperparameter tuning which uses an approximate model of the system dynamics and backpropagation to directly optimize hyperparameters for the closed-loop DeePC performance. Numerical simulations demonstrate the effectiveness of DeePC in combination with DeePC-Hunt in a complex stabilization task for a nonlinear system and its superiority over model-based control strategies in terms of robustness to model misspecifications.} }
Endnote
%0 Conference Paper %T DeePC-Hunt: Data-enabled Predictive Control Hyperparameter Tuning via Differentiable Optimization %A Michael Cummins %A Alberto Padoan %A Keith Moffat %A Florian Dorfler %A John Lygeros %B Proceedings of the 7th Annual Learning for Dynamics \& Control Conference %C Proceedings of Machine Learning Research %D 2025 %E Necmiye Ozay %E Laura Balzano %E Dimitra Panagou %E Alessandro Abate %F pmlr-v283-cummins25b %I PMLR %P 673--685 %U https://proceedings.mlr.press/v283/cummins25b.html %V 283 %X This paper introduces Data-enabled Predictive Control Hyperparameter Tuning via Differentiable Optimization (DeePC-Hunt), a backpropagation-based method for automatic hyperparameter tuning of the DeePC algorithm. The necessity for such a method arises from the importance of hyperparameter selection to achieve satisfactory closed-loop DeePC performance. The standard methods for hyperparameter selection are to either optimize the open-loop performance, or use manual guess-and-check. Optimizing the open-loop performance can result in unacceptable closed-loop behavior, while manual guess-and-check can pose safety challenges. DeePC-Hunt provides an alternative method for hyperparameter tuning which uses an approximate model of the system dynamics and backpropagation to directly optimize hyperparameters for the closed-loop DeePC performance. Numerical simulations demonstrate the effectiveness of DeePC in combination with DeePC-Hunt in a complex stabilization task for a nonlinear system and its superiority over model-based control strategies in terms of robustness to model misspecifications.
APA
Cummins, M., Padoan, A., Moffat, K., Dorfler, F. & Lygeros, J.. (2025). DeePC-Hunt: Data-enabled Predictive Control Hyperparameter Tuning via Differentiable Optimization. Proceedings of the 7th Annual Learning for Dynamics \& Control Conference, in Proceedings of Machine Learning Research 283:673-685 Available from https://proceedings.mlr.press/v283/cummins25b.html.

Related Material