On the certainty-equivalence approach to direct data-driven LQR design

F Dörfler, P Tesi, C De Persis - IEEE Transactions on Automatic …, 2023 - ieeexplore.ieee.org
The linear quadratic regulator (LQR) problem is a cornerstone of automatic control, and it
has been widely studied in the data-driven setting. The various data-driven approaches can …

Logarithmic regret bound in partially observable linear dynamical systems

S Lale, K Azizzadenesheli, B Hassibi… - Advances in Neural …, 2020 - proceedings.neurips.cc
We study the problem of system identification and adaptive control in partially observable
linear dynamical systems. Adaptive and closed-loop system identification is a challenging …

Active learning for nonlinear system identification with guarantees

H Mania, MI Jordan, B Recht - arxiv preprint arxiv:2006.10277, 2020 - arxiv.org
While the identification of nonlinear dynamical systems is a fundamental building block of
model-based reinforcement learning and feedback control, its sample complexity is only …

Non-asymptotic and accurate learning of nonlinear dynamical systems

Y Sattar, S Oymak - Journal of Machine Learning Research, 2022 - jmlr.org
We consider the problem of learning a nonlinear dynamical system governed by a nonlinear
state equation ht+ 1= ϕ (ht, ut; θ)+ wt. Here θ is the unknown system dynamics, ht is the …

Sample complexity of linear quadratic gaussian (LQG) control for output feedback systems

Y Zheng, L Furieri, M Kamgarpour… - Learning for dynamics …, 2021 - proceedings.mlr.press
This paper studies a class of partially observed Linear Quadratic Gaussian (LQG) problems
with unknown dynamics. We establish an end-to-end sample complexity bound on learning …

Active learning for nonlinear system identification with guarantees

H Mania, MI Jordan, B Recht - Journal of Machine Learning Research, 2022 - jmlr.org
While the identification of nonlinear dynamical systems is a fundamental building block of
model-based reinforcement learning and feedback control, its sample complexity is only …

Learning mixtures of linear dynamical systems

Y Chen, HV Poor - International conference on machine …, 2022 - proceedings.mlr.press
We study the problem of learning a mixture of multiple linear dynamical systems (LDSs) from
unlabeled short sample trajectories, each generated by one of the LDS models. Despite the …

Online learning of the kalman filter with logarithmic regret

A Tsiamis, GJ Pappas - IEEE Transactions on Automatic …, 2022 - ieeexplore.ieee.org
In this article, we consider the problem of predicting observations generated online by an
unknown, partially observable linear system, which is driven by Gaussian noise. In the linear …

Smoothed online learning for prediction in piecewise affine systems

A Block, M Simchowitz… - Advances in Neural …, 2023 - proceedings.neurips.cc
The problem of piecewise affine (PWA) regression and planning is of foundational
importance to the study of online learning, control, and robotics, where it provides a …

Adaptive control and regret minimization in linear quadratic Gaussian (LQG) setting

S Lale, K Azizzadenesheli, B Hassibi… - 2021 American …, 2021 - ieeexplore.ieee.org
We study the problem of adaptive control in partially observable linear quadratic Gaussian
control systems, where the model dynamics are unknown a priori. We propose LQGOPT, a …